Metacognition and Modeling in Chemistry Instruction

Kennesaw State University
DigitalCommons@Kennesaw State University
Doctor of Education in Secondary Education Department of Secondary and Middle Grades

Dissertations Education
Fall 11-29-2021
Metacognition and Modeling in Chemistry Instruction

Meri Cain
Follow this and additional works at: https://digitalcommons.kennesaw.edu/seceddoc_etd
Part of the Science and Mathematics Education Commons
Recommended Citation
Cain, Meri, "Metacognition and Modeling in Chemistry Instruction" (2021). Doctor of Education in
Secondary Education Dissertations. 32.
https://digitalcommons.kennesaw.edu/seceddoc_etd/32
This Dissertation is brought to you for free and open access by the Department of Secondary and Middle Grades
Education at DigitalCommons@Kennesaw State University. It has been accepted for inclusion in Doctor of
Education in Secondary Education Dissertations by an authorized administrator of DigitalCommons@Kennesaw
State University. For more information, please contact digitalcommons@kennesaw.edu.
Running head: METACOGNITION AND MODELING IN CHEMISTRY INSTRUCTION 1
by
Meri Laird Cain
A Dissertation
Presented in Partial Fulfillment of Requirements for
the Degree of Doctor in Education
In
Secondary Chemistry Education
In the
Bagwell College of Education
Kennesaw State University
Kennesaw, Georgia
November 29, 2021

METACOGNITION AND MODELING IN CHEMISTRY INSTRUCTION 2
By
Meri Laird Cain
Approved:
___________________________________________________________
Dr. Mei Lin Chang

(Committee Chair)
___________________________________________________________
Dr. Michelle Head

(Committee Member)
___________________________________________________________
Dr. Rachel Gaines

(Committee Member)
___________________________________________________________
Dr. Adrian Epps

Dean
Bagwell College of Education
Acknowledgements
I would like to thank my family for encouragement and support throughout this journey. I
am grateful to my mother, Cornelia Laird and my father, Dr. Earl Laird, for instilling in me self-
discipline, a strong work ethic and the drive to achieve. I am grateful to my siblings, Jennifer
Bartholomew, Sara Irby, Kathleen Hogue and Lee Laird, for cheering me on to this
accomplishment. I could not have completed this arduous undertaking without the constant
support and kind forbearance of my children, Ferrill Clark, Taylor Cain, Jarred Cain, Kathryn
Ray, Nathan Cain and Harrison Cain. It is they who have endured with me in the grind of the
journey and spurred me on with their gracious patience and inspiring rally.
I am ever grateful to the leadership and the uplifting support of my dissertation
committee chair, Dr. Mei Lin Chang; she was a bright light beckoning me on at every juncture of
the path and bringing expertise to the rescue as needed. Dr. Michelle Head, also on the
committee, started me down this road and believed I could walk it with success; she set before
me challenges that deepened my understanding of two complex topics. Dr. Rachel Gaines, also
on my committee, was an immense and providential source of guidance during the uphill climb
of the dissertation. To each one, who fulfilled their roles with grace and excellence, I am
grateful.
This work is dedicated to my children, Ferrill, Taylor, Jarry, Kathryn, Nate and Harry,
because the raising of my children made me into a person who could take on a difficult goal late
in life and finish it. May you each become fully who you were created to be. Thanks be to God
for grace and goodness to set before me a new path and give me strength, perseverance and
capability to walk it. To Your glory, I have earned new letters.

Abstract
The purpose of this research was to examine the interplay between science modeling, a
science practice, and direct instruction in metacognition. The quasi-experimental study was
designed to investigate how students’ skill in science modeling was bolstered by the
fortification of increased metacognitive skill and how science modeling and metacognitive
guidance augmented the acquisition of chemistry content. Further, perceptions of
metacognitive value and of metacognitive thinking processes were appraised in conjunction with
the direct metacognitive cultivation treatment. Data was collected from 48 students in honors
chemistry class at a large suburban high school in the Southeastern United States (control group,
n=27; experimental group, n=22). Students in the experimental group were introduced to
metacognition through direct instruction and taught modeling skills alongside metacognitive
reflection; thus, direct cultivation of metacognition was contextualized in the development of
modeling skill. Achievement data and survey on perceptions of the metacognitive thinking
process were collected. The results showed no significant differences between control and
experimental groups in the survey nor for learning gains from the overall achievement data. On
the other hand, significant differences between groups were revealed for performance measures
of rubric scores on models and summative scores on model construction. A positive relationship
was revealed between value of metacognition and metacognitive thinking processes for the
experimental group. The results of the research are significant for chemistry teachers in the skill
development of modeling and potential in bolster metacognition.
KEYWORDS: chemistry modeling, particulate drawings, submicroscopic
representations, SEPs, solution chemistry modeling, metacognition, direct instruction of
metacognition, contextualized metacognition in content skills

Table of Contents
Acknowledgements ......................................................................................................................... 3
Abstract ........................................................................................................................................... 4
Chapter One: Introduction .............................................................................................................. 8
Background and Rationale .......................................................................................................... 9
Definition of Terms ................................................................................................................... 13
Purpose of Study ....................................................................................................................... 16
Scope and Limitations ............................................................................................................... 17
Significance of the Study .......................................................................................................... 17
Chapter Two: Literature Review .................................................................................................. 19
Constructivism in Science Instruction ...................................................................................... 19
Conceptual Framework ............................................................................................................. 20
Scientific Inquiry in Science Education .................................................................................... 20
Models and Modeling ................................................................................................................ 24
Metacognition............................................................................................................................ 28
Metacognition in Science Instruction ........................................................................................ 34
Metacognition Observed in the Chemistry Classroom of the Researcher ................................ 39
Survey Instrument Chosen for Adaptation. ............................................................................... 41
Theoretical Framework ............................................................................................................. 42
Research Questions ................................................................................................................... 43

Chapter Three: Methodology ........................................................................................................ 46
Setting........................................................................................................................................ 48
Participants ................................................................................................................................ 49
Instrumentation.......................................................................................................................... 50
Reliability and Ethical Considerations ...................................................................................... 55
Data Analysis ............................................................................................................................ 63
Anticipated Results ................................................................................................................... 67
Effect of Direct Instruction of Metacognition on Modeling ...................................................... 67
Effect of Direct Instruction of Metacognition and Modeling on Learning Gains ..................... 67
Chapter Four: Results ................................................................................................................... 70
Confirming Equivalence Before Intervention ........................................................................... 70
Research Question One: Metacognition Enhancement ............................................................. 71
Metacognition Enhancement of Modeling ................................................................................ 71
Metacognition Enhancement of Content Acquisition ................................................................ 77
Research Question Two: Direct Instruction of and Awareness of Metacognition .................... 79
Research Question Three: Value and Awareness Correlation in Experimental Group ............ 81
Chapter Five: Discussion .............................................................................................................. 83
Implications of Findings and Connections with Prior Literature .............................................. 87
Limitations of the Study ............................................................................................................ 92
Future Research ......................................................................................................................... 94

Implications to the High School Chemistry Classroom ............................................................ 95
References ..................................................................................................................................... 96
Appendix ..................................................................................................................................... 113

Chapter One: Introduction
When I embarked on Masters’ level of study at the University, I was not a novice
instructor rather a veteran of 20+ years in the classroom. Little did I know that two concepts
acquired under the tutelage of two masterful professors would form the tandem areas of study
foundational to my doctoral study. Learning about the potential of metacognition and the power
of modeling forcefully impacted my instructional practices. When I stood at the precipice of
determining research topics for doctoral study a few years later, these topics, transformational in
my paradigm of teaching, would call. Metacognition and modeling could be joined together in
the research purpose. The design of the study would be reminiscent of the enjoyment found in
the quantitative statistics, encountered for the first time in graduate school, that I labored to learn.
Hence, I view the three precursors supporting the final research study to have been the important
and the influential learning that took place when I, myself, was a graduate student – in my early
fifties.
The need for this study is founded on the research accomplished in the past in the areas of
science modeling, direct metacognitive instruction and the benefit of contextualizing
metacognition into content skills. Science modeling and metacognition are topics with much
research available to inform instruction, however, studies demonstrating integration of science
content skills with metacognition are limited. This study sought to fill that gap by advancing
understanding of contextualizing the development of metacognition within the specific content
skill of modeling. The rationale for the study is founded on the need for research to example a
pedagogical strategy fusing metacognition and modeling which in turn strengthens our
understanding of cultivating metacognition and of developing the science skill of modeling. The
purpose of the chapter at hand is to frame the topics of study and to introduce the focus of the
research to study the interplay between metacognition and modeling in chemistry.
Background and Rationale
The goals of the more recent Georgia Standards of Excellence science standards embrace
the procedural understanding of science produced through the Science and Engineering Practices
(SEP), prescribed by the NRC (National Research Council, 2012), and promote the role of
metacognition in learning (Dooley, 2017). The SEPs, key process skills highlighted in recent
national guidance of science instruction, could be strengthened by well-developed skills in
metacognition and, at the same time, carrying out the science practices could increase student
development of metacognition. A study of metacognitive cultivation contextualized in the
development of one of the SEPs, could prove to be impactful research for the community of
science teachers in Georgia and beyond.
Science teachers are tasked with developing science practices in our students in an
integrated fashion with our science content. “Developing and using models” is SEP #2 (National
Research Council, 2012). This practice is embedded into the Georgia Standards of Excellence,
GSE, science standards throughout the curriculum. Although the advantage of practicing
chemistry modeling has been studied and endorsed (Cooper et al., 2017; Justi & van Driel, 2006;
Leenars et al., 2013 Louca & Zacharia, 2012; Sujak & Daniel 2017; Swartz, et al., 2017), the
practice of using and creating models is difficult for students (Khan, 2011). Additionally,
recognition of the role of models and modeling is recent in science education so many science
teachers are not equipped to teach skillfully from a modeling perspective (Justi & van Driel,
2006). Evidence in the literature supports a study on the acquisition of modeling skill since the
literature tells us that chemistry teachers and chemistry students have problems with
understanding the nature of models and modeling as a practice (Erduran & Mugaloglu, 2014;
Khan, 2011; Windschitl et al., 2008). Carpenter et al. (2019) noted that modeling is one practice
that teachers struggle to fully understand and implement despite the practice being primary in the
work of science.
Although the research community has accomplished much work regarding the effect of
modeling on conceptual understanding of chemistry, it has produced little research regarding the
students’ cognitive processes involved in learning via modeling (Louca & Zacharia, 2012).
Research strongly indicates the need for metacognitive expertise to develop knowledge of
science through inquiry methods of science (Hacker, 2009) such as those found in the SEPs.
Evidence has shown that students with higher metacognitive capabilities show greater gains
when learning via inquiry-based instruction methods, such as modeling, than do the students
with lower metacognitive capabilities (Hacker, 2009). This suggests that enhancing
metacognition would lead towards augmenting the development of modeling skills.
Given that modeling is very challenging, introducing the practice to students during a
modeling exercise produces metacognitive experiences, which refer to a student’s awareness and
feelings elicited in a problem-solving challenge. Development of metacognitive thinking
processes, prompted by metacognitive experiences, have the potential to strengthen modeling
practice and could be reflected in enhanced student mastery of the conceptual content served by
the modeling. The success of intertwining metacognitive instruction with modeling will serve to
help teachers and students alike for whom modeling practice is initially a challenge (Carpenter et
al., 2019).
The science practice of modeling could be strengthened by well-developed skills in
metacognition and, at the same time, carrying out the modeling practice increases skills in
metacognition. The pedagogical strategy of this study to offer direct guidance in metacognition
in tandem with developing one of the SEP examined the effect of pairing modeling with
metacognition; the combination was unique and offered gains for student success in chemistry.
The process of creating a model, by drawing a particulate representation of a process or
structure, was connected to conscious reflection in metacognition when direct guidance in
metacognition was done in tandem with the modeling activity in chemistry. The science practice
of modeling and metacognition held the promise to benefit each other in a reciprocal fashion and
to contribute to the acquisition of learning.
Instruction in metacognitive skills has been demonstrated to provide a more student-
centered environment in the chemistry classroom, to facilitate metacognitive processes and to
promote self-efficacy in chemistry (Kirbulut, 2014). According to Schraw, et al., (2006) the
critical role metacognition can serve in scientific inquiry has motivated scrutiny of the balance of
time devoted to teaching concepts in science in the absence of metacognitive instruction. It was
noted to be better to devote the time to developing procedural understandings of science.
Procedural skills, such as science modeling in chemistry, are considered linked to metacognitive
skills (Schraw et al., 2006). Some science educators propose increasing time for procedural
learning because procedural competence, linked to metacognition, is a large allotment of the
problem solving and critical thinking demanded in higher levels of science education (Schraw et
al., 2006).
The topic of instruction in metacognition is vital to understanding the role of
metacognitive thinking in learning. A key debate noted by Zohar and Barzilai (2013) is whether
metacognition is domain-general or domain-specific. The domain-specific view of metacognition
is supported by studies (Cook et al., 2013; Schraw, 1995; Veenman, 2013) where metacognition
is contextualized within specific tasks, specific domains, and specific content areas. The direct
instruction of metacognition delivered within a specific domain (i.e., science) and the matrix of a
content-specific skill (i.e., science modeling) holds the promise of potent development of
metacognitive practices in students.
Students who “gain a level of awareness above the subject matter” (Chick, 2013, p.1)
expand learning strategies most effectively when metacognitive instruction is embedded into the
content and the activities with which they are engaged (Chick, 2013). Direct instruction of
metacognition and contextualizing metacognitive cultivation within the SEP of modeling is a
novel application and a beneficial example of how to embed metacognitive practice within a
specific content. Therefore, the research study explored metacognitive instruction embedded into
an inquiry-based SEP utilized in chemistry instruction to gather understanding useful to the
community of chemistry educators.
This research examined an explicit contextualization: that of direct metacognitive
instruction and guidance applied to the demanding chemistry practice of modeling. The evidence
documented through this research, to support integration of the metacognitive instruction of into
a content-specific task, furthers the current understandings of how to cultivate metacognitive
practices in our students.
The value of direct metacognitive instruction has been established in conjunction with a
positive effect on achievement in chemistry (Cook & McGuire, 2013; Kingir & Aydemir, 2012;
Rahman et al., 2010; Uopasai et al., 2018). Specific facets of chemistry acquisition, such as
problem solving (Cooper & Sandi-Urena, 2009), self-efficacy (Kirbulet, 2014; Zubeyde, 2014),
and concept learning (Thomas & Anderson, 2014; Thomas & McRobbie, 2013; Tanner, 2012)
have each been explored in concurrence with metacognition. Likewise, the learning advantage of
practicing chemistry modeling has been studied and endorsed (Cooper et al., 2017; Justi & van
Driel, 2006; Leenars et al., 2013 Louca & Zacharia, 2012; Sujak & Daniel 2017; Swartz et al.,
2017).
This research fills a gap in the existing research by integrating metacognitive instruction
and chemistry modeling, two evidence-based instructional pedagogical strategies that have yet to
be examined in tandem. This research is positioned to contribute unique findings about
metacognition and modeling. It can strengthen appreciation for the reciprocal benefit of
metacognitive instruction and content-specific contexts in science and, indeed, all content areas.
Definition of Terms
Scientific Inquiry
Scientific inquiry describes the approaches of problem-solving and the processes of
scientists used in investigating and explaining natural phenomenon by diverse methods (Klar,
2000). Schawrtz et al. (2004) defines scientific inquiry as characteristics of the processes by
which scientific knowledge is developed. Scientific knowledge and scientific theories are built
through scientific inquiry. In establishing a working definition of scientific inquiry, various
definitions found in the literature (Akuma & Callaghan, 2018; Chen et al., 2020; Cullen, 2015;
Mupira & Ramnarain, 2017; White et al., 2009) were synthesized and tailored to the goals of this
research. This process produced the following definition:
Scientific inquiry originates with a question, posed by teacher or student, wherein the
answer is sought by collecting evidence to support the construction of an explanation. The
explanation is communicated, undergoes critique and revision before becoming a refined,
justifiable answer to the question.

The research purported to investigate the effect of an independent variable, a cognitive
intervention, on dependent variables, content skill and concept acquisition, to answer the
question of how contextualizing metacognitive cultivation could enhance the science practice of
modeling and the learning of chemistry. Hence the research in its very design overarchingly
corresponded to the definition of scientific inquiry. The definition of scientific inquiry supports
the experience of the participants in the study since a content question was investigated by
evidence observed in the phenomena and an explanation was constructed by students in the form
of a model which was critiqued, revised and refined to answer the question.
Metacognition
Metacognition is cognition that goes beyond commonplace thinking (Rahman et al.,2010)
and is often defined as “thinking about one’s thinking” (Cooper & Sandi-Urena, 2009) since the
term was first coined by John Flavell in the 1970s (Kinger & Ayedmir, 2012). Schraw and
Dennison (1994) included both knowledge of cognition and regulation of cognition within the
definition of metacognition. Yen et al. (2018) described metacognitive knowledge,
metacognitive skills, and metacognitive experiences as a part of self-regulated learning. MK is
the declarative knowledge about the interplay between person, task and strategy (Veenman,
2006). For example, a student knowing that memorization of element symbols is an easy task for
her but that learning to work chemistry problems will take more work is MK. MS is the
conscious and purposeful application of strategies and regulation of thinking to achieve desired
outcomes. Examples of MS include a chemistry student taking careful notes on problem
examples and referencing those notes as he works on the homework problems or a student
making flashcards to memorize element symbols. It includes regulation, for example, a student
checking a posted key, after completing homework, to correct errors. ME are the feelings or
judgement experienced by the learner when a new learning challenge is presented. ME can also
serves as feedback to aid in regulation of MS (Yen et al., 2018).
The definition of metacognition that was used in this study, offered by Thomas and
McRobbie (2013), is an “individuals’ knowledge, control and awareness of cognition” (p. 302)
representing MK, MS and ME.
Models and Modeling
Models in chemistry education are diagrams or drawings which represent a view of a
system, a phenomenon or process. Models often represent the sub-microscopic or the particulate
level of Johnstone’s triangle in chemistry (Thomas & McRobbie, 2013). Modeling is the work of
a student to create a labeled or keyed drawing which represents the unseen elements present in a
system, process or phenomena (Cooper et al., 2017). Students engaged in modeling observe or
reflect on a macro process or event and interpret unseen processes and activity. The procedure
involved in this research study replicated model-based inquiry as the basis for answering a
specific conceptual question with an explanation (Louca & Zacharia, 2011). Synthesizing and
refining a definition for models and for modeling based on the literature (Cooper et al., 2017;
Louca & Zacharia, 2011; Santos & Arroio, 2016; Thomas & McRobbie, 2013). and tailored to
the goals of the research accorded the following definitions:
Models in science explain and predict through representation something in the natural
world; models show how an observed phenomenon works.
Modeling is the construction and revision of a representation to communicate the
characteristics and/or relationships between components of a system with explanatory and
predictive capability.
The use of a model to represent and communicate internalized understandings reflects the
collaborative nature of science discovery in the work of scientists. The protocol of refining a
model equally reflects the journey of science to refine knowledge of the universe through
critique, testing and revision. Scientists communicate, collaborate and refine explanations to
deepen and validate the understandings we hold about our world. Modeling, therefore, is a
microcosm mirroring the mission of scientific inquiry.
Phenomenon
Phenomenon in chemistry instruction is an observable event ideally interesting, complex
and aligned to the standard under study (“The Wonder of Science”, n.d.). It can be
demonstration of a physical process or a chemical reaction conducted as a demonstration. It can
take the form of a video or a simulation showing the same also. Talanquer (2018) explained how
students need to be able to investigate and make sense of macroscopic phenomena, tangible and
visible events, to interpret chemical systems and to make arguments regarding causes.
Purpose of Study
The purpose of this research was to examine the interplay between science modeling and
direct instruction in metacognition. The quasi-experimental study was designed to
investigate how students’ skill in science modeling was affected by the fortification of
increased metacognitive skill and how each strategy, science modeling and metacognition,
augmented the acquisition of chemistry content. Further, perceptions of metacognitive value
and of metacognitive thinking processes were appraised in conjunction with the fortification of
metacognitive skill. The research questions considered aspects of integrating metacognition

instruction into the context of content skills for the novel application of metacognitive
assimilation into the specific SEP of modeling.
The following research questions guided this study:
1. Does direct coaching or instruction of metacognition enhance modeling and
the acquisition of the conceptual content?
2. Does direct coaching or instruction of metacognition enhance student awareness of
metacognitive thinking processes?
3. Is there a correlation between student perception of the value of metacognitive
activities and their awareness of metacognitive thinking processes?
Scope and Limitations
The research was conducted in the honors chemistry courses led by the researcher. Thus,
it was considered quasi-experimental research. The risk of backyard research includes the study
not being viewed as seriously by others. The process of the research was thoroughly scrupulous
and data analysis handled with meticulous detail to garner respect of fellow educators.
Participants in the study were students in the honors level chemistry course and thus it is
convenience sampling. The school is a large suburban high school located outside of a large
metropolitan area in the Southeast where the majority of the students are from affluent families
with high SES.
Significance of the Study
Improving teaching and learning from evidence-based findings is the goal of researchers
in chemistry education (Towns, 2013). The robust benefit of designing an activity to serve
modeling, metacognition and conceptual learning was exciting and useful to improve chemistry
education. Contextualizing metacognition into content-specific skills added another layer of
significance embedded in this study.
The significance of research in chemistry education should be evaluated on criteria
including novelty, the impact and the influence of the findings (Towns, 2013). Investigating a
research question that possessed novelty, had potential for impact and to exert influence in the
field of chemistry education seemed daunting, yet this study built on the interplay between
modeling and direct metacognitive instruction could benefit the field of chemistry education and
could exert a far-reaching effect on the teaching practices of many chemistry teachers.
Chapter Two: Literature Review
Constructivism in Science Instruction
The perspective on learning undergirding this study emanates from model-based inquiry
instruction which is grounded in the constructivist tradition (Louca & Zacharia, 2012). Scientific
inquiry engages students in an authentic learning process to discover the workings of the natural
world (Fox & Risconscente, 2008; Uopasai et al, 2018). Students can construct and internalize
real meaning through their own investigations and build an experience-based understanding of
why we believe what we believe about science (White et al, 2009).
Constructivism and Modeling
As described by Uopasai et al., (2018), the constructivist philosophy of science teaching
and learning is centered on student mental models and their misconceptions. This possesses
important implications for teachers who aspire to demonstrate scientific reasoning for their
students. The teaching model, based on constructivism, draws on learning theories pioneered by
Piaget, Vygotsky and James (Fox & Risconscente, 2008; Uopasai et al., 2018). Specifically, the
model emphasizes the critical need to establish a cognitive framework and use relevant information
to propel conceptual change.
According to Thomas and McRobbie (2013), student reasoning in chemistry should take
place though conscious consideration of chemical phenomena at the macroscopic, the
molecular/sub-micro and the symbolic levels and should generate construction, or revision, of
the learners’ conceptual beliefs based on evidence. Within scientific inquiry, as evidence leads
toward an explanation, the explanation can be communicated, critiqued and revised via the
evaluation of models or within the practice of modeling to refine the answer to a scientific
question and to strengthen the justification of the answer proposed (Louca & Zacharia, 2012).
The scientific practice of modeling unequivocally undergirds and enhances the progression from
the unknown to the known in the process of scientific inquiry.
Constructivism and Metacognition
Conscious, learner-regulated construction is at the heart of cognitive development (Glaser,
1984) and involves awareness of thinking and learning, which are key elements in metacognition
(Thomas & McRobbie, 2013). The metacognitive approach of students using prior knowledge to
plan a strategy, implement it, reflect upon it, evaluate the results and modify the approach plays a
critical role in successful (Uopoasai et al., 2018).
Conceptual Framework
Support in the literature for the current study was provided within the conceptual
framework. Specifically, the conceptual framework laid out three pillars of scientific inquiry: an
understanding of models and modeling in science, the importance of metacognitive skills in
chemistry learning and the benefit of direct instruction in metacognition contextualized into
content skill for student success.
Scientific Inquiry in Science Education
Historical Evolution of Scientific Inquiry. Scientific inquiry, at its first enumeration in
education, was proposed in contrast to the view of science before 1909, when science was
understood to be a body of knowledge to be learned through direct instruction (National
Research Council, 2000). In 1909, John Dewey argued for a change to the instruction of science
to include the process or methods of scientists (Dewey, 1910). Dewey’s model defined six steps:
sensing perplexing situations, clarifying the problem, formulating a tentative hypothesis, testing
the hypothesis, revising with rigorous steps and acting on a solution (Barrow, 2006).
Science instruction gained national focus after the 1957 launch of Sputnik I. In 1960,
Joseph Schwab encouraged science teachers to teach in a manner like science operates (Barrow,
2006). Schwab emphasized integrating scientific inquiry into science instruction by looking to
the laboratory and using those experiences to lead instruction rather than as exercises to follow
instruction (National Research Council, 2000). Schwab developed four variations of scientific
inquiry in the classroom where the role of the student becomes increasingly independent: (1).
questions are posed by the teacher and methods are provided to determine answers, (2). questions
are posed by teacher and methods are open for students to determine, (3). students determine
questions and methods and propose explanation based on experimentation, and (4). students
conduct ‘enquiry into enquiry’ by using published experiments to establish explanations
(National Research Council, 2000). By the 1970s, there were growing expectations that scientific
inquiry should be a part of teaching science so that students would become involved in doing
science rather than only being told about or reading about science (National Research Council,
2000).
The National Science Education Standards (NSES), a landmark public policy document,
was issued in 1996 and expanded the idea of scientific inquiry in the classroom to include
learners using inquiry steps and knowing the steps of conducting scientific inquiry (Barrows,
2006). This was followed by the publication of Inquiry and the National Science Standards,
which identified five essential features of inquiry as follows: (a) scientifically oriented questions,
(b) evidence collected by students, (c) explanations developed by students, (d) evaluation of their
explanations (e) communication and justification of their explanations (National Academies of

Sciences Engineering and Medicine, 2000). Unpacking the NSES to address inquiry, three
domains of inquiry in education were described to include the abilities to develop in students, the
understanding of the scientific method students should know, and teaching standards for the
inclusion of scientific inquiry.
Current Understanding of Inquiry. The original pedagogical term, inquiry, has evolved
into “inquiry-based learning” (Heindl, 2019), which refers to the infusion of scientific inquiry
into the classroom. This approach is now seen as essential for gaining scientific knowledge in the
classroom as it has always been in pursuit of scientific knowledge among scientists. Inquiry-
based learning departs from traditional learning because the teacher is not the center of the
process. Thus, students are engaged in inductive discovery as opposed to being led in recipe
science (Heindl, 2019). Inquiry strategies change the focus of instruction from learning what we
know about science to learning why we believe what we believe about science. Thus, students
gain understanding based on experience (Cullen, 2015).
Chen et al. (2020) note that existing literature offers different viewpoints about the
fundamentals of scientific inquiry in the classroom. The essential abilities or competencies based
on a synthesis of the literature on scientific inquiry are questioning, predicting, investigating,
interpreting data, explaining, and communicating (Chen et al., 2020). Akuma and Callaghan
(2018) describe inquiry-based learning as a reflection of scientific inquiry consistent with the
social constructivist learning perspective which emphasizes knowledge being actively
constructed by the learner. Mupira and Ramnarain (2017) concur that inquiry-based learning is
constructivist, inductive and active learning characterized by questioning, data-analysis and
critical thinking to construct real meaning and knowledge. Inquiry-based learning is meant to
engage learners in an authentic discovery process (Pedaste et al., 2015, as cited by Mupira &
Ramnarain, 2017).
Scientific inquiry has been viewed as a process of oscillating between theory and
evidence within the practice of competitive argumentation (White et al., 2009). Scientific inquiry
leads toward the development of scientific laws, models and theories by proposing theories then
seeking evidence through investigation (White et al., 2009). From this perspective, the basic
model of scientific inquiry is described as: (a) theorizing, (b) questioning and hypothesizing, (c)
investigating and (d) analyzing and synthesizing (White et al., 2009).
Science Practices. More recently, there is a new emphasis driven by the K-12
Framework for Science Education (National Research Council, 2012) on mirroring the scientific
practices of scientists and to illustrate the nature of science by shifting from the terminology of
inquiry-based teaching methods to the assimilation of authentic scientific practices (French &
Burrows, 2018). Authentic Scientific Inquiry (ASI) is defined as a fuller version of inquiry
teaching that aligns more closely with the work of real scientists in contrast with traditional
classroom labs (French & Burrows, 2018). NRC, in the 2012 public policy publication for U.S.
national science standards, defined the scientific process as “practices” designed to engage
students in authentic scientific inquiry and define the ways scientists gain knowledge (Donohue
et al., 2020).
The practices, formally entitled Science and Engineering Practices, consist of the
following: (a) Asking Questions and Defining problems, (b) Developing and Using Models, (c)
Planning and Carrying Out Investigations, (d) Analyzing and Interpreting, Data (e) Using
Mathematics and Computational Thinking, (f) Constructing Explanations and Designing
Solutions, and (g) Engaging in Argument from Evidence (National Research Council, 2012). The
SEPs can be implemented with tiered strategies reflecting different levels of inquiry-based
instruction. The SEPs may be integrated using traditional methods (i.e., question given→method
given→answer given). However, the integration of the SEPs with the three highest inquiry
levels, described by Schwab (1962) as “question open→method open→answers open”, affords
the greatest inclusion of the SEPs (Akuma & Callaghan, 2018). Therefore, open inquiry-based
strategies align most with accomplishing the SEPs and affiliate strongly with scientific inquiry of
science in the most authentic manner (Akuma & Callaghan, 2018).
Models and Modeling
Models as Metaphors. Models in science are explained by Schwartz and White (2005)
as a set of representations, rules and reasoning structures that allow for the generation of
prediction and explanation. Models can be considered extended metaphors that serve to apply
information and understanding from the source domain, that of our common experience, to the
target domain, the typically abstract and new concept we seek to master (Brown, 2003). Models
as extended metaphors can guide thinking about a science system under investigation, however,
they can also inhibit thinking about the system in other ways. According to Brown (2003), the
role metaphor and modeling can play in successful science education is highlighted as it
complements the sister goals of imparting conceptual understanding and generating a sense of
intellectual excitement about the subject. These tools can correct embodied ideas that students
bring with them and develop scientific reasoning. One of the pillars in explaining the
construction of a metaphor include grounding it in common experiences with the physical world
and understandings from the social domain (Brown, 2003).
Model Construction. Model-based learning in science occurs via student construction of
models that represent physical phenomenon and externalize the underlying mechanism of an
unseen process (Louca & Zacharia, 2012). Modeling is the process of building, using and
evaluating external representations of systems to make sense of phenomena (Swartz et al.,
2017). In the science classroom, viewing physical phenomenon and participating in group
discussion pave the way for fruitful student - modeling (Leenaars et al., 2011). Gray and Rogan-
Klyve (2018) summarized modeling as a reasoning practice used to anchor the “complex work of
understanding and implementing scientific practices in the classroom to explain phenomenon
and solve problems” (p. 1346). According to the NGSS (NRC, 2012) “Models serve the purpose
of being a tool for thinking with, making predictions and making sense of experience” (p. 56).
For scientists, models represent current understandings, help identify questions and explanations
and are a means of communication. For that reason, “Developing and Using Models” is one of
the SEPs recommended in the K-12 Framework for Science Education (National Research
Council, 2012). Models include the following: diagrams, physical replicas, mathematical
representations, analogies, and computer simulations (NRC, 2012).
Models are considered not merely representations of science processes and facts but are
“tools for reasoning” (Swartz et al., 2017, p. 114). A modeling approach in science education is
undergirded by constructivist learning theory (Leenaars et al., 2011). Leenaars et al. (2011)
describe modeling construction as following the construction of internal mental models and
manifesting perceptions. When the learner constructs an external model, it makes it concrete and
moves it into the social dimension of science. Cooper et al. (2017) concluded, based on their
research, that sketching to construct a science model is shown to be vital for revealing the quality
of the student’s mental models of a chemical process.
Models in Chemistry. The importance of models in the construction of chemical
knowledge is suggested by philosophy of chemistry according to Erduran and Mugaloglu (2014).

Visual models are a dominant way of thinking in the subject of chemistry (Akaygun & Jones,
2014). The use of modeling is considered vital in explaining the relationships between the
macroscopic (what can be seen) and the microscopic/particulate levels (Santos & Arroio, 2016).
Representational Levels in Chemistry. Learning chemistry with understanding involves a
conscious inter-relating of three representations of chemistry: the macroscopic (what can be
observed), the molecular/microscopic (what is happening on the unseen level), and symbolic (the
abstract, written communication, such as an equation) as described by Talanquer (2011). The
chemistry ‘triplet’ of representation was first proposed by Johnstone (1993). The emergence of a
recognition and enumeration of three components of chemistry in common use provided the
chemistry instructors a manageable paradigm to use (Talanquer, 2011).
Models and Representational Levels. Proficiency in the use of these multiple
representations, macroscopic, submicroscopic and symbolic, in chemistry has been demonstrated
to show improved understanding for grade nine students when studying chemical reactions and
that the submicroscopic, in particular, must be described with models although it is the most
difficult to nurture (Chandrasegaran et al., 2011). Within the submicroscopic representational
level, the science practice of particulate modeling is utilized as an explanatory means of showing
that representational level. Particulate modeling exhibits understanding and facilitates a bridge
between macroscopic phenomenon and submicroscopic representation (Santos & Arroio, 2016).
Metacognition and Representational Levels. Louca and Zacharia (2012) note that
making the connections between the three representations emphasizes metacognition because the
process engages students to consider how they think about chemistry. Thomas and Anderson
(2012) have presented similar findings on the role of metacognitive awareness in enhancing
students’ ability to move from one representation of chemistry to another.

Learning Through Model Construction. Learning through modeling, as described by
Louca and Zacharia (2012), emerges from the interaction of the learners’ experiences and ideas
which are communicated and solidified through construction of an artifact, a concrete model.
This process, called constructive modeling (Kokkonen, 2017), is often embedded in an inquiry
activity wherein a model is constructed to generate a hypothesis about a question or problem
regarding a phenomenon. Constructed models are developed, tested and revised within the
inquiry activity. Cooper et al. (2017) defined modeling as “sketching the invisible to predict the
visible,” which develops model-based reasoning and is an important means of assessing
understanding in the chemistry classroom (p. 902). Since modeling is considered a skill critical
to developing the molecular/microscopic conceptual understanding (Louca & Zacharia, 2012), it
is a vital tool.
Model Construction and Misconceptions. Sujak and Daniel (2017) contended that the
non-emphasis of chemistry teachers on the molecular/microscopic level leads to misconceptions.
Modeling can promote accurate understanding and help identify misconceptions (Santos &
Arroio, 2016). Furthermore, Cooper et al. (2017) found that students who drew more models
showed the greatest learning gains and these drawn models were invaluable as assessment tools
for revealing students’ misconceptions.
Model Construction and Increased Content Acquisition. When Kimberlin and Yezierski
(2016) studied the effect of high school students engaged in using particulate level modeling in
stoichiometry, they found the intervention significantly improved conceptual understanding of
stoichiometry. Moreover, they found anecdotal evidence of gains in algorithmic understanding of
stoichiometry. Similarly, Okumus et al (2019) found that the combination of cooperative
learning and creation of models increased conceptual understanding about chemical reactions in
a general chemistry course populated by pre-service science teachers. In a study conducted by
Edwards and Head (2016), when modeling was paired with inquiry in an activity introducing the
vocabulary of the representational levels in chemistry, students showed gains in understanding
and application of the vocabulary.
Metacognition
Metacognition, a term originating from the work of John Flavell (1979) and Ann Brown
(1987), describes one’s knowledge about and regulation of one’s cognitive abilities in learning
processes. Metacognition has been catch-phrased as ‘thinking about thinking” (Chick, 2013).
The researchers Schraw et al. (2006) align metacognition alongside cognition and motivation as
equal components of learning.
Metacognitive Theory. Metacognition is considered a facet of cognitive learning theory
and an application of social-cognitive learning theory, For the psychologist James, metacognition
was an activity of the self. He paralleled awareness of one’s own cognition with maturing into
adulthood when such thinking was thought to become habitual (Fox & Riconscente, 2008).
Piaget defined metacognition as conscious awareness and the ability to communicate rationale
for one’s thinking (Fox & Riconscente, 2008). Vygotsky’s work expanded the definition of
metacognition to include not only awareness of cognitive activity of the mind but also control
and direction of thoughts by use of signs. Vygotsky also described the knowledge of one’s
mental powers in relation to a given task as another key facet of metacognition (Fox &
Riconscente, 2008). According to Vygotsky, metacognition is enhanced by scientific concepts
and promoted through social interaction with others, including adults. The use of language, via
social interaction, and the acquisition of scientific concepts promote MK and propels the child
toward a mature ability to grasp and formulate concepts, and to transfer new understandings to
novel situations (Fox & Riconscente, 2008).
Components of Metacognition. Metacognition can be divided into two broad,
distinctive and independent components: a) knowledge of cognition and b) metacognitive
regulation (Schraw et al., 2006; Deng et al., 2011). Metacognition has also been defined as three
aspects of a cognitive process: knowledge of the process, monitoring of the process and control
of the process (Hacker, 2009). Bowen et al. (2017) agree that metacognition is not only
knowledge of how one learns or thinks but also has a regulatory aspect involving planning,
monitoring and evaluating. Planning, monitoring and evaluating are prescribed steps to
consolidate the skill of understanding and monitoring one’s learning or cognitive processes
(Bowen et al., 2017). Regulation of cognition is applied to the exact task and thus includes
“planning for, monitoring the progress of and evaluating projects or investigations” (Hacker,
2009, p.180).
Definitions of Metacognition. After the inception of the term metacognition by Flavell
(1979), many different “flavors” of definitions were proposed, some of which are redundant and
others conflicting (Zohar & David, 2009). Veenman et al. (2006) noted the domain of
metacognition lacks coherence and has inconsistency in its concepts although its importance is
agreed upon. The literature suggests, as noted in examples above, that the definition of
metacognition is usually three parts with synonymous although non-identical components. The
three components articulated by Tobias and Everson (2002) seem to represent this accepted
structure wherein metacognition can be broken down into knowledge about metacognition,
ability to monitor the learning process and the meta-ability to control the learning process.
Scholars (Thomas & McRobbie, 2013; Yen et al., 2018) have divided metacognition into
three facets: metacognitive knowledge (MK), metacognitive skills (MS) and metacognitive
experiences (ME). Thomas and McRobbie (2013) define metacognition as “individuals’
knowledge, control and awareness of cognition” (p. 302) representing MK, MS and ME. MK is
declarative knowledge about the interplay between person, task and strategy (Veenman, 2006);
MS is the conscious and purposeful application of strategies and regulation of thinking to
achieve desired outcomes and ME is the feeling or judgement experienced by the learner which
also serves as feedback to aid in regulation of MS (Yen et al., 2018).
More recently, Graham et al. (2019) detail the three facets of metacognition:
(1) Metacognitive knowledge/awareness: an individual’s knowledge of what they know
and/or beliefs about their own self, knowledge of the task, goals, and knowledge of
various effective strategies. (2) Metacognitive skills/strategies: the ability to think about
what you are currently doing and then select and utilize appropriate learning strategies
such as planning, self-monitoring, self-evaluation, self-regulation, orienting, and more.
They reflect the skill with which an individual can monitor, guide, and regulate their own
thinking and learning. (3) Metacognitive experiences: the affective (emotional) or
cognitive experiences that occur during metacognition (p. 1539).
Under metacognitive knowledge/awareness falls the components of metacognitive knowledge
(declarative) and metamemory. Under metacognitive skills/strategies falls metacognitive
knowledge(conditional) and metacognitive skills. The third facet is metacognitive experiences. It
is also noteworthy that metacognitive beliefs affect declarative knowledge, procedural
knowledge and metacognitive experiences (Graham et al., 2019.)

Metacognitive Declarative Knowledge. Distinct facets or components of metacognition
have each been highlighted in the literature. A common distinction in metacognition that appears
in the literature separates MK or declarative knowledge from MS or procedural knowledge
(Schraw, 1988; Thomas & McRobbie, 2001; Veenman et al., 2006). Declarative knowledge is
one’s beliefs about the tasks, conceptions about one’s abilities and encompasses one’s goals;
while procedural knowledge (part of MS) refers to knowledge about how one performs cognitive
tasks or how one regulates problem-solving.
Metamemory, which is associated with developing declarative knowledge (Veenman et
al., 2006), is often studied in younger children and is concerned with the knowledge and control
of one’s memory (Karably & Zabrucky, 2009). It contains two components: stable knowledge
(i.e., how one learns including understanding of the person, task and strategy of learning), and
procedural metamemory, which emanates from meta-experiences where memory becomes
instrumental in recalling how to use strategies and in evaluating progress (Karably & Zabrucky,
2009).
Metacognitive Skills. MS entail procedural knowledge, conditional knowledge and
meta-strategic knowledge (MSK). The components described in metacognition as conditional
metacognitive knowledge recognize the value and the limitations of procedural knowledge
(Thomas & McRobbie, 2001) and are about knowing “what to do when” (Veenman et al., 2006,
p. 5). Declarative and procedural knowledge can serve to equip students, but it is the conditional
knowledge that appropriately applies the two making metacognition effective. All three interact
in the function of metacognition (Thomas & McRobbie, 2001) although the interaction of the
three is not yet clearly defined (Zohar & David, 2008). Specifically, in the context of science
inquiry, White and Frederiksen (2005) defined MK and MS in three parts as: (a) knowledge
about capabilities and goal structures needed in science inquiry, (b) knowledge of how one
organizes and manages science inquiry processes and (c) knowledge of how to apply one’s
metacognitive capabilities in science inquiry.
MS have been spotlighted as advantageous for all learners in the literature. Veenman et
al. (2006) demonstrated that metacognitive skills, uniquely account for 17% of variance in
learning, intellectual ability accounts for 10% of variance in learning and the two together share
another 20% of variance in learning for students of different ages and background, for different
types of tasks, and for different domains. The implication is that proficiency in metacognition
may compensate for students’ cognitive limitations (Veenman et al., 2006). White and
Frederiksen (2005) report findings of responses from students, who participated in metacognitive
development through computer-guided activities, which indicated that students saw their
capabilities as subject to reflection and improvement. A significant decrease in the performance
gap between high-achieving and low-achieving students was linked to the metacognitive
approach by the researchers thus encouraging the “important idea that anyone can learn and
improve” (White & Frederiksen, 2005, p. 222).
Metastrategic knowledge, MSK, is a term linked to procedural and conditional
knowledge of metacognition (Kuhn, 2001; Zohar & David, 2008). MSK is defined as general
knowledge about higher order thinking strategies (Zohar & David, 2009) that involves
understanding a thinking strategy and knowing when it should be used along with being able to
outline the strategy. MSK also incorporates an understanding of the task as part of meta-strategic
thinking (Kuhn, 2001; Zohar & David, 2008).
Metacognitive Experiences. ME are believed to be critical to the development of
metacognitive knowledge by becoming the basis for metacognitive reflection (Thomas &
McRobbie, 2013). ME are experiences related to cognitive endeavors or to behaviors or to
metacognitive knowledge which elicit monitoring of cognitive process or outcomes (Thomas &
McRobbie, 2013). Experiences are the feelings and the judgements taking place during
metacognition (Graham et al., 2019). ME are most powerful when contextualized within
appropriate content-rich activities (Crick, 2013; Thomas & McRobbie, 2013). These experiences
create an opportunity for students to reflect and revise as they are prompted to think about how
they learn in relation to the content area and to evaluate the consequences of changing learning
processes (Thomas & McRobbie, 2013). ME produce feedback to behavioral control processes
as students monitor and assess outcomes thus allowing for deficiencies to be corrected (Graham
et al., 2019).
Metacognitive Beliefs. Metacognitive beliefs describe beliefs held about one’s thinking
(Flavell, 1979) and are considered to have a place in self-regulation and in judgements made in
ME (Nelson et al., 1999). Metacognitive beliefs act as a declarative knowledge about the self and
will prompt action to self-regulate when an individual desires a state different than the one they
are in presently (Nelson et al., 1999). Positive metacognitive beliefs relate to the perceived
potential of ‘thinking’ to benefit an outcome. It is noted that even test anxiety can be considered
positive metacognitive belief because the individual thinks that worrying about a test will help
them be more successful on the test. Negative metacognitive beliefs are based on believing that
there is danger in ‘thinking’. An individual, for example, may believe that thinking about bad
things could cause bad things to happen (Fergus et al., 2020).
In summary, metacognition is both about oneself and about the current, precise mission
of learning being undertaken. The metacognitive approach of students (using prior knowledge to
plan a strategy, implement it, reflect on and evaluate the results, and modify the approach as
called for) plays a critical role in successful science learning (Uopoasai et al., 2018). Regulation
of cognition goes hand-in-hand with procedural understanding of science which can be learned
through inquiry and the science practices (Hacker, 2009). Investigation into their own
metacognitive practices is a path to enable students to learn about scientific theorizing and
inquiry while also increasing their metacognitive growth (Hacker, 2009).
Metacognition in Science Instruction
The literature reveals manifold approaches to investigating metacognition in science and
in chemistry education specifically. The value of embedded metacognitive instruction has been
established in conjunction with a positive effect on achievement in chemistry (Deng et al., 2011;
Thomas & Anderson, 2014; Thomas & McRobbie, 2001; Uopasai et al., 2018). Deliberate
instruction in metacognitive strategies in the science classroom appears to enhance both
metacognition and learning in a broad range of students (Cook &McGuire, 2013; Graham et al.,
2019; Seraphin et al., 2012; Zepeda et al., 2015). Use of interventions to produce metacognitive
experiences have been explored in chemistry instruction (Casselman & Atwood, 2017; Cooper &
Sandi-Urena, 2009; Thomas & McRobbie, 2013) and metacognitive awareness has been
examined for relationships to chemistry student efficacy and achievement (Kingir & Aydemir,
2012; Kirbulet, 2014; Rahman et al., 2010).
Direct Instruction in Metacognition. Direct, purposeful instruction and specific
guidance about metacognition and metacognitive strategies have been examined in the research
literature and specifically in science education. Three fundamental principles described by
Veenman et al. (2006) for successful metacognitive instruction involve: (a) embedding
metacognitive instruction in the content to ensure connectivity, (b) informing learners about the
usefulness of metacognitive activities to encourage the initial extra effort, and (c) prolonged
training.
Approaching instruction through constructivist practices featuring integrated
metacognitive guidance has proven beneficial. Uopasai et al. (2018), for instance, found a
significant difference in overall academic outcomes between experimental and control groups of
veterinary medicine students when the experimental group was subject to a teaching intervention
based on constructivism, metacognition and neurocognition as compared with traditional
teaching methods in the control group. The teaching method was credited with enhancing
students’ abilities in understanding medical terminology and anatomical knowledge, and with
increasing metacognitive ability, including declarative, procedural and conditional knowledge
(Uopasai et al., 2018). Similarly, Deng et al. (2011) compared content acquisition and
metacognition in eleventh grade chemistry students in China. When the treatment group was
given a constructivist data-logging activity (as opposed to the traditional teacher-centered
methods used in the control group), both metacognition and conceptual understanding increased.
Studies have illustrated how deliberate use of language that emphasizes and guides
metacognition in the chemistry classroom improves student outcomes as well. For example,
Thomas and McRobbie (2011) found that embedding a constructivist metaphor “learning is
constructing” into the daily work of Australian chemistry students generated variable effects on
student metacognition. The case study, featuring a catalytic ME, showed some gains in MK
specifically (Thomas & McRobbie, 2001). More recently, Thomas and Anderson (2014)
confirmed that changing the learning environment of a chemistry classroom through the use of
language that explicitly addresses metacognition and uses metaphor to guide how students think
about the three representational levels of chemistry can lead to changes in students’
metacognitive knowledge. The Metacognitive Orientation Learning Environment Scale- Science
(MOLES-S; Thomas, 2003) was used alongside the Self-Efficacy, Metacognition Learning
Inventory- Science (SEMLI-S; Thomas et al., 2008) with additional support of student interviews
(Thomas and Anderson, 2014). The study highlights the idea that students’ metacognition should
be developed and enhanced through teaching activities which embed training in metacognition
into science learning (Thomas & Anderson, 2014).
Short Term Instruction. Short exposure to direct instruction of MS or metacognitive
strategies is also supported in the literature as having an impact on academic performance, the
growth of student metacognition, the conceptual understanding of chemistry and self-efficacy
beliefs of students toward chemistry. The following referenced studies support the methodology
of short exposure to direct metacognitive instruction for significant gains in chemistry. Cook et
al. (2013) found that through a mere 50 minutes of direct instruction about metacognition,
college chemistry students without metacognitive learning strategies can be taught metacognitive
strategies resulting in significant improvement in test scores in chemistry. Specifically, students
in the study were instructed on MS and introduced to a study cycle eliciting ME. According to
surveys and interview, attending a lecture on metacognition was shown to change student
behavior resulting in a statistically significant improvement in test scores (Cook et al., 2013). In
another study examining the effect of direct instruction on metacognition, Zepeda et al. (2015)
implemented a six-hour intervention implemented designed to teach declarative and procedural
metacognition. Science students performed significantly better on a conceptual test and a self-
guided learning activity after exposure to the direct instruction on metacognition.
Longer Term Instruction. There have also been studies of longer-term interventions
based on direct instruction of metacognition. Graham et al. (2019), for example, developed an
approach for direct instruction of metacognition framed in weekly chemistry tutor sessions in
college introductory chemistry courses. These sessions produced gains in self-efficacy in STEM
and increased student achievement. Development of MK and MS were the focus of the twelve
tutoring sessions followed by an opportunity to apply MK and MS directly to chemistry tasks
therefore inducing ME (Graham et al., 2019). Seraphin et al. (2012) developed an instructional
intervention that involved explicit teaching of metacognitive strategies and metacognitive
reflection coupled with science inquiry, which they presented to teachers enrolled in professional
development over two years’ time. When the teachers implemented these practices with students,
they were shown to increase science processing skills and metacognition in both teachers and
students. These examples demonstrate effects of long-term instruction in metacognition for
science and chemistry education.
Instruction Based on ME in the Chemistry Classroom. Metacognitive guidance in
conjunction with an inquiry activity produces a learning environment rich in ME in which
students can develop metacognition and self-efficacy in chemistry (Kirbulut, 2014). Researchers
who develop an intervention to draw out ME, have seen gains in metacognition and achievement
in chemistry. Based on interviews with students conducted after the students participated in an
inquiry activity, Bowen et al. (2017) found that the students credited the activity with prompting
metacognition. Further, the students described intermediate or high-level MS in their actions
during the activity. This research suggests that even students are cognizant of the interaction
between metacognition and inquiry, as well as, the resulting benefits when they reflect on
learning activities (Bowen et al., 2017). Casselman and Atwood (2017) stimulated ME by means
of a treatment group experiencing direct metacognitive training in chemistry online homework
exercises (i.e., where students predicted quiz scores, received the quiz feedback and followed up
with self-designed weekly study plans to address the quiz results). The metacognitive
intervention produced significant gains in test performance over the course of a semester with the
lowest quartile of students seeing the greatest gains (Casselman & Atwood, 2017).
Sandi-Urena et al. (2011) promoted ME in chemistry classrooms via a collaborative
intervention that simulated a cognitive imbalance experience to provoke metacognitive reflection
and social interaction. An increase in use of MS, a significant increase in metacognitive
awareness, as indicated by the Metacognitive Activities Inventory (MCAI; Cooper & Sandi-
Urena, 2009), and an increase in problem-solving abilities in chemistry were all gains supported
by the intervention (Sandi-Urena et al., 2011). Thomas and McRobbie (2013) utilized an
interpretive methodology in a study with multiple data sources from a treatment group emanating
from a teachers’ pedagogical shift to an activity system to elicit ME and guided reflection for
two years. Videos of class time and interviews provided data in interviews to validate the effect
of a changed pedagogy to develop student metacognition (Thomas & McRobbie, 2013).
Student Awareness of Metacognitive Practices. Metacognitive awareness has been
highlighted in the literature as a promising indicator for producing success in science
performance. Rahman et al. (2010) investigated the role of metacognitive awareness in academic
performance in chemistry and found significant correlation between students’ metacognitive
awareness and their performance on a researcher-made chemistry test. The study measured
metacognitive awareness using an inventory, Schraw and Dennison’s Metacognitive Awareness
Inventory (MAI; 1994), and measured performance using a researcher-created test and
subsequently conducted a correlational analysis (Rahman et al., 2010). The metacognitively
aware students performed well on the test demonstrating that awareness of metacognition can
hold a profound potential for improving performance in chemistry.

In addition, the findings of a study conducted by Kinger and Aydemir (2012) reveal a
strong positive association between attitudes toward chemistry and metacognition awareness.
The study compared the MAI (Schraw & Dennison, 1994), and Attitude Scale toward Chemistry
in conjunction with student course averages to reveal significant correlations between the three.
Additional information was obtained via a Student Background Questionnaire containing
questions pertaining to topics such as parental education, resources (i.e., books, study desk)
available at home and parental employment. The study noted the students possessed greater
declarative knowledge than procedural knowledge in chemistry indicating room for further
integration between procedural knowledge and metacognition (Kinger & Aydemir, 2012). These
findings suggest that strengthening procedural knowledge is a desirable goal in a learning
activity. In an investigation of the relationship between the self-efficacy of students and
metacognitive awareness, Kirbulut (2014) found that highly efficacious students were more
aware of their own knowledge, (i.e., MK), and regulation, (i.e., MS), of cognitive processes.
Kirbulut’s (2014) action research data suggests that specifically in the domain of chemistry,
highly efficacious chemistry students were more aware of their metacognition.
Corresponding to the above studies, metacognitive awareness in chemistry education has
been demonstrated to produce gains in chemistry learning. The strong recommendation based on
the study of Rahman et al. (2010) is that metacognitive strategies be encouraged in students
through instructional activities since significant impact on student performance was seen for
students with higher metacognitive awareness.
Metacognition Observed in the Chemistry Classroom of the Researcher
Formative Observations
In the chemistry classroom, MK can be used and further developed as students encounter
all aspects of chemistry learning. When students encounter quizzes and tests, they must draw
upon MK about themselves, tasks and strategies. This is evidenced by the increase in capability
on quizzes after the first few quizzes. The students acquire a know-how in approaching this new
kind of math-integrated science and with representational forms previous unbeknownst to them.
Consequently, the metacognitive declarative knowledge specific for the study of chemistry
appears to grow. Conditional metacognitive knowledge of when to apply a strategy or an
approach becomes important in assessments, daily work and lab work. The students coming in
from biology, where the strategies for success are different, always seem to have an adjustment
when they enter the chemistry classroom. For example, students know that daily work is only
graded for completion and shown work, not accuracy. However, many students after they
complete homework, go up to the front of the room to check their work against the handwritten
key. By early in the semester, it is a minority of students that do not review the posted key and
thus it seems most have recognized by conditional MK a new strategy to apply. Likewise,
students develop strategies and procedures to ensure lab report success. They learn that all
conclusions need actual data. MS is seemingly employed to recognize the task performance that
will produce the desired outcome when a lab is scored and utilize the strategies needed. Quizzes
are formative assessments that provide opportunities to grow in metacognitive thinking skills
because they alert the student to areas of concern, affecting MK, and prompt a self-regulation,
MS, before the test to adapt. Students develop and apply the strategies needed to become
successful when they receive feedback and are made aware of not reaching a desired goal.
Students seem to grow in metacognition through quizzes.
Summative Observations
A discussion of metamemory and metacognitive beliefs must include summative
assessment behavior observed in the classroom. By year 10 and 11 of high school, students have
much familiarity with how their individual memory works and what helps them to master a
content area before a summative evaluation. Also, they have metacognitive beliefs about review,
cramming, coming to WEB review, making flash cards, highlighting notes, or showing up as
soon as possible to sit for a test. One sees diverse approaches to preparing for tests that are
informed by their positive metacognitive beliefs about actions that will help them succeed. One
also observes that there are some students who claim little to no preparation will help them
indicating an established metacognitive belief relating to tests. For tests, students do not seem to
have adapted strategies to the task at hand for the specific content but they usually comment on
reliance on well-established methods founded on long held metacognitive beliefs.
Survey Instrument Chosen for Adaptation.
The instrument Metacognitive Activities Inventory, MCAI, (Cooper & Sandi-Urena,
2009), was chosen for adaptation by the researcher. MCAI was found to be a robust, reliable and
validated assessment of metacognition in chemistry problem solving (Cooper & Sandi-Urena,
2009). The researchers described the findings in the validation study:
Reliability was measured in terms of internal consistency, as well as the
reproducibility observed after retesting. Validity was examined in two dimensions: face
validity, in terms of the acceptability and reasonableness to those who are tested; and
construct validity, in the extent to which its items conform to the functional definition of
the construct measured and its ability to predict group differences. It is estimated that the
evidence gathered sufficiently supports the validity of the inventory.” (Cooper & Sandi-
Urena, 2009, p. 244).

The MCAI was used in a study as a data collection method to assess and strengthen an approach
to measuring metacognition in chemistry problem solving using multi-methods (Cooper et al.,
2011). An additional study featuring an intervention of providing cooperative, project-based
laboratory experiences to increase problem-solving ability and metacognitive strategies also
utilized the MCAI (Cooper et al., 2012).
Theoretical Framework
Constructivism Undergirds Learning Chemistry through Modeling
Constructivist learning theory applied to the teaching of science is said by Colburn
(2000) to involve the refining of student beliefs about science to become more in line with the
beliefs held by the scientific community. He explains that science teaching from a constructivist
perspective will guide students to understand “how and why scientifically accepted explanations
explain and predict what will happen in a given situation better than their intuitive ideas” (p. 10).
As knowledge is actively constructed in an inquiry method via the modeling of a chemistry
phenomenon, misconceptions can be replaced with solid scientific ideas about how the world
works. Research strongly indicates the need for metacognitive expertise to develop knowledge of
science through inquiry methods (Hacker, 2009).
Metacognition and Modeling Interplay to Benefit the Learning of Chemistry
Metacognition is considered an application of social-cognitive learning theory and
metacognition is a facet of cognitive theory. MK encompasses our specific knowledge of the
content at hand or knowing “what you know and what you don’t know” about the specific
content to be mastered. MS and the regulation of cognition are applied to the exact task and thus
includes “planning for, monitoring the progress of and evaluating projects or investigations”
(Hacker, 2009, p.180). Therefore, the student benefits from metacognition which is defined by
Veenman (2011) as “knowing what one knows and does not know, from knowing oneself in the
learning process, from reviewing and evaluating choices and evidence” (p. 213).
Planning, moderating and evaluating are requisite mechanisms for student metacognition
(Bowen et al., 2017). The practice of modeling elicits active engagement from the student and
produces a metacognitive experience. Guiding student reflection on the three processes of
metacognitive practices in tandem with an activity of science modeling promotes skill in
metacognition. Science practices are strengthened by developing skills in metacognition and, at
the same time, carrying out the practices increases skills in metacognition. The SEPs and
metacognition benefit each other in a reciprocal fashion.
The purpose of this research was to examine the interplay between science modeling and
direct instruction in metacognition. The quasi-experimental study was implemented to
investigate how high school chemistry students’ skill in science modeling could be bolstered by
the fortification of increased metacognitive skill and if each strategy, science modeling and
metacognition, could augment the acquisition of chemistry content. It investigated what impact
integrating metacognitive instruction into modeling activities had on the acquisition of the
chemistry content addressed by the model constructions. See the summary of the framework
illustrated in Figure 1.
Research Questions
The following research questions guided this study:
1. Does direct coaching or instruction of metacognition enhance modeling and the acquisition
of the conceptual content?

3. Is there a correlation between student perception of values of metacognitive activities and
their awareness of metacognitive thinking processes?

Chapter Three: Methodology
Rationale for Design
The research method was conducted as a quasi-experimental design utilizing quantitative
analysis; a quasi-experimental design seeks the nature of the relationship between an
independent variable and a dependent variable. Quasi-experimental does not possess a true
experimental design because the participants are not randomly assigned to the groups of control
group and experimental group. The design methodology was appropriate to the natural setting of
the classroom and was an effective means of demonstrating the effect of the intervention or
independent variable, although, the methodology could expose threats to validity and limitation
in generalizing the results. The choice of quasi-experimental design utilizing quantitative data
and analysis is appropriate for the positionality of the researcher of post-positivism wherein
cause and effect relationships can be observed to discover truth.
The present study was driven by the constructivist’s theory of learning, based on the
assumption that learners are active participants in the learning process, and thus metacognition
could be cultivated through direct instruction. Therefore, the study involved integrating an
intervention (i.e., direct instruction in metacognition) into chemistry content. The goal of the
intervention was to enhance learning by applying direct instruction of metacognition to
modeling. The investigation into possible correlational relationships between cultivation of
metacognition, the science practice of modeling, and the possible influence on student
performance in chemistry signaled the use of quantitative research methods. The effect of an
intervention was best assessed through experimental or quasi-experimental design, which
incorporates independent and dependent variables. Groups that are matched in characteristic (i.e.,
age, course level, content expectations, teacher) comprise fitting samples for the experimental
and control groups.
The research design reflects the theory that metacognition integration into content is
powerful. Cultivating metacognition is best, not as a stand-alone strategy, but when instruction
of metacognition is immersed into a content skill; it is best embedded into the activities about
which students are thinking (Chick, 2013). Zohar and David (2007) demonstrated teaching
metacognition within a concrete learning experience is valuable because students use a thinking
strategy within a context and are not merely learning metacognition in abstract. The placement of
metacognition within development of modeling offered a theoretical confirmation of the work
of Zohar and David (2007) who showed that metacognitive instruction is most impactful when
contextualized within content.
Design
The study sought to understand if students perceived favorably the value of direct
metacognitive instruction to enhance their science skills, and if the direct instruction of
metacognition improved the science skill of modeling and student acquisition of the content.
In answering the first research question on the effect of direct coaching/ instruction of
metacognition on modeling and the acquisition of conceptual content, the independent variable
was whether or not students received the intervention (i.e., direct instruction on metacognition).
Specifically, the independent variable was defined as a presentation on the concept of
metacognition and use of a metacognitive guide for reflection during the introduction and
subsequent practice of modeling. The dependent variables were measured with scores on the
models produced and learning gains scores based on the scores of the pretest/posttest. To
underscore the investigation of the effect of metacognitive guidance on the science practice of
modeling, the course unit test items evaluating the model construction were also compared.
In answering the second research question on how direct coaching/instruction for
metacognition affects students’ awareness of metacognitive processes, the independent variable
was whether or not students received the intervention (i.e., direct instruction on metacognition)
and the dependent variable were scores on the survey instrument.
In answering the third research question on if there is a relationship between student
perception of value of metacognition and students’ awareness of metacognitive processes, the
two variables were compared using scores on the survey instrument.
In summary, the dependent variables were defined as scores on models, learning gains
scores from pretest/posttest and scores on models from the course unit test alongside the
responses to student survey on metacognitive activities. The instructor, the age group, and the
level of chemistry were controls since both experimental and control groups will share these
same characteristics. The control variables, class of students and level of chemistry course, are
defined as sophomores or 10th grade and honors level students, and the intervening variables,
academic motivation and gender are defined as baseline test average and female/male.
Setting
The research was conducted in large suburban high school located outside of a major
metropolitan area in the Southeast. It has been recognized nationally as a high-achieving school
of excellence, ranked among the top 150 schools in the nation by U.S. News and World Report,
and has been ranked consistently among the top five in the state (US News and world Report,
2020). The school demographics were described in 2020-2021 as 64% white, 20% Asian, 6%
Black, 7% Hispanic, 3% multiracial and other (K-12 Public Schools Report Card, 2021). The
gifted population was 44%, the graduation rate was 94.6% in 2021 and 4% of the student body is
considered economically disadvantaged, meaning they are eligible for free or reduced lunch (K-
12 Public Schools Report Card, 2021).
The student population has an overall characteristic of an unusual awareness of their
academic progress and goals. The gifted population is 44% of the school community. School
attendance is high, in part, due to a previous incentive policy. The data for the three years
previous to the Covid-19 pandemic had attendance for all students at 66.1% for 5 or fewer
absences; most of the 66.1% were absent only 2 days per semester (K-12 Public Schools Report
Card, 2019). The incentive policy has been suspended for the current school year as a response to
the Covid-19 pandemic. Students attended school either face-to-face or virtually during the same
class period during the semester of the research.
Participants
Convenience sampling was used, specifically, all students from the honors chemistry
classes currently instructed by the researcher were invited to participate in the study (n = 76).
The ages of the sophomore students ranged from 15-16 years and the students were
predominantly sophomores and juniors in high school. The student population in the chemistry
courses was roughly the same as the school population. Students attended class during the same
time period each day but attended either face-to-face or virtually.
Data was collected from those granted permission to participate by informed consent and
student assent forms distributed prior to the lesson cycle (see APPENDIX A). As this is a quasi-
experimental study, the sample was not randomly assigned to control and experimental groups
but was assigned according to course section. Seventy-eight students were invited to participate
in the study who were enrolled in three honors course sections. All students experienced the
activities: the two smaller sections (n=46) did not have the treatment whilst the larger third
section (n=29) had the treatment. Data was collected from those students who submitted consent
and assent forms for the control group (n=27) and the experimental group (n=22). For the control
group participants, 20 students attended virtually and 7 attended face-to-face; for the
experimental group, 16 students attended virtually and 6 attended face-to-face.
The identities of students were concealed by assigning project identification numbers on
all collection items throughout the data collection. All documents, pretests/posttests, models and
Metacognitive Maps, were labeled with the project identification number assigned for the study.
Every student was assigned a number and was required to submit assignments, however, only
students who had submitted permission forms from parents and consent forms for themselves
had data entered for use in the study. Project identification numbers were used throughout the
data analysis.
Instrumentation
Instrumentation and data sources include a metacognitive question guide (Metacognition
Map), a scoring rubric to evaluate models, a pre-test and post-test of a chemistry concept, and the
Metacognitive Activities and Perceptions Inventory (MCAPI). Each of these measures will be
described in the follow sections.
The Metacognition Map
The Metacognitive Map instrument was developed by the researcher for use in the
current study (see APPENDIX B). The questions within the Metacognition Map were founded
on the work of Tanner (2012) who patterned types of questions to guide students toward
metacognitive activity. The question types suggested by Tanner were adapted and placed into the
format of a question guide for use alongside the modeling activities. The guiding questions
prompted responses associated with the three regulatory tasks of metacognition identified by
Bowen et al. (2017): planning, monitoring and evaluating. The Metacognition Map was used to
enhance metacognitive instruction to the experimental group when students completed it
alongside the construction and revision of five models.
Achievement Scores
Student achievement data was based on (a) pre-test and post-test, each comprised of 22
questions, on the misconceptions identified in the literature related to the concepts demonstrated
by the chemical phenomenon (see APPENDIX C) and (b) five rubric scores from evaluation of
the modeling exercises (see APPENDIX D for rubric and APPENDIX E for model instructions
as appeared in the Formative app) and also (c) scores for the models constructed during the
course unit test (see APPENDIX F for unit test questions). The specific pre-test/post-test was
designed to evaluate misconceptions in the content area of solution chemistry; likewise, the
phenomena that undergird the model constructions were selected to be appropriate for the
content in solution chemistry.
Development of Pretest/Posttest Instrument. The performance measure to determine
learning gains of students in the concepts of solution chemistry was based on specific
misconceptions identified in the literature. The pretest and posttest were designed to target the
misconceptions found in the literature about solution chemistry and the construction of the
questions was sourced from sources having established reliability and validity as national tests. A
written version is found in APPENDIX C however, the version delivered to students was
delivered online via Formative app.

Adadan and Savasci (2012) noted alternative conceptions about the nature of solutions
and dissolving in water such as solutes melting before dissolving into water or turning into liquid
before mixing with water and even that solutes react with water. Calyk et al. (2005) describe the
same misconception and note that particulate drawings and diagrams help students to express the
correct concepts with more success. The pretest questions # 1-3 address these misconceptions
with one question adapted from online quiz (Queensland Science Teachers, 2000) the second
from an exam review (New Jersey Center for Teaching and Learning, 2014) and the final is a
researcher generated question using a particulate drawing, representing the sub-microscopic level
and equations representing the symbolic level.
Nakhieh (1992) reported student misconceptions of relating chemical equations to correct
particulate drawings of a sub-microscopic system, (i.e., dissolving), and question # 3 addresses
the appropriate equation for dissolving associated with the particulate representation of
dissolving. Misconceptions regarding the dissolving of ionic substances have been documented
and researchers have found that students struggle to understand the separation into ions by ionic
compounds (Kelly et al., 2010; Naah & Sanger, 2012), have difficulty transferring the dissolving
of the exemplar NaCl to other ionic salts (Devetak et al., 2009) and often fail to apply subscripts
in ionic compound formulas to the number of ions produced in solution (Smith and Metz, 1996).
Nakheih (1992) and Kabapanar et al. (2004) denote student difficulty with matching dissolving
equations with submicroscopic representations and correctly depicted a hydration ring. Questions
# 4-12 are about ions or electrolytes dissolving and were sourced from Noah and Sanger (2012),
College Board (n.d.) multiple choice and Regents (2018, 2019).
Questions # 11 (Adadon & Savasci, 2012) and # 12, sourced from American Chemical
Society Chemistry Olympiad (2012) are about saturation of solute in solution. Adadan and
Savasci (2012), Pinarbasi and Canpolat (2003), and Devetak et al. (2009) each addressed the
misconceptions with understanding and representation of saturated solutions and supersaturated
solutions noting that students will represent both types of solutions with identical drawings. The
questions addressing gasses dissolved into solutions, # 13-16, concern misconceptions about gas
solutes noted by Adadan and Savasci (2012). Colligative properties for solid and gas solutes are
evaluated in questions # 17-22 where the inability to distinguish pure solvents and solutions by
properties, a noted misconception, are examined (Adadan & Savasci, 2012; Pinarbasi &
Canpolat, 2003). These questions, # 13-22, are sourced primarily from College Board (n.d.),
Regents (n.d.), and American Chemical Society (n.d.) Chemistry Olympiad.
Rubric for Models. The modeling activity was scored using a rubric designed to
evaluate the success of the model to communicate concepts evidenced in the chemical
phenomenon (see APPENDIX D.). This evaluation rubric holds face validity and content validity
emanating from its development by the researcher and two other colleagues, each veteran
chemistry instructors, for the purpose of scoring student-made models in chemistry (Edwards et
al., 2017). The rubric holds criterion validity since it was based on models exampled in the
literature and cited at the foot of rubric including Merritt and Krajcik (2013), Merritt et al.,
(2008), Ngai et al. (2014) and, Sevian and Talanquer, (2014). The rubric addressed features of
the models such as illustration of particles, use of spacing and/or size, use of motion/energy,
particle interaction, number of particles depicted and use of levels of representation. Evaluation
of the student models will supply the performance data on the quality of the student-created
models.
The rubric used has reliability established by experts locally at the school through use by
veteran chemistry instructors. Human error in evaluating the models using the rubric might
become an issue and was minimized by having two instructors grade 5% of the models. Also,
researcher bias, such as performance bias or measurement bias, could be introduced in the
grading of the models thus another evaluator in addition to the researcher scored the rubrics to
circumvent that possibility.
MCAPI. The third instrument, the MCAPI was used to collect data on the development
of metacognition and students’ perception of metacognition (see APPENDIX G). The MCAPI
was adapted from the Metacognitive Activities Inventory (MCAI; Cooper & Sandi Urena, 2009)
for this study by the researcher. The MCAI is a robust, reliable and validated assessment of
metacognition in chemistry problem solving and has been shown to exhibit acceptable levels of
internal consistency, reproducibility, face validity and construct validity for a college population
(Cooper & Sandi-Urena, 2009). The MCAPI instrument was derived from the MCAI and
modified to address the specific skill of modeling activity.
Adaptation of Survey Instrument for Study. Although the MCAI was tested with a
college population, the reliability and validity of the MCAI is expected to translate in the MCAPI
since both groups, college and honors chemistry undertake the same chemistry skills in
modeling. The MCAI items on problem solving were adapted to address the different chemistry
skill, modeling. MCAPI is comprised of 28 statements. Agreement with the items is indicated on
a Likert scale from 1 (strongly disagree) to 6 (strongly agree). The first six items address student
perceptions and were devised by the researcher. The subsequent twenty-two questions mirror the
questions on the MCAI (Cooper & Sandi-Urena, 2009), but apply to modeling instead of to
problem-solving in chemistry. Six of those questions are negatively phrased and reverse scored.
Subscales of MCAPI. The MCAI was adapted into the Metacognitive Activities and
Perception Inventory, MCAPI, to fit the goal of two subscales: that of perception of value for
metacognitive thinking and that of perception of metacognitive thinking processes applied to the
skill of modeling. Within the second subscale of perception of metacognitive thinking skills,
three components were present aligning with the Metacognitive Map, as follows: Planning,
Moderating and Evaluating. These components reflect the three regulatory tasks of MS identified
by Bowen et al. (2017) and used by Tanner (2012). See APPENDIX H for MCAPI organized by
subscales and components. Reliability for the MCAPI was assessed; Cronbach alpha for subscale
one measuring perception of value of metacognitive thinking, was .80 and Cronbach alpha for
subscale two measuring perception of metacognitive thinking activities in modeling, was .90.
The subscale one of the MCAPI, concerning value of metacognitive thinking, was
administered to the experimental group only in order to obtain feedback on student perceptions
regarding the value of metacognition. The subscale two of the MCAPI measured the perception
of metacognitive activities in modeling and was administered to the experimental group and the
control group after the lesson cycle. Qualtrics was utilized to administer the survey and collect
the data via the online platform of the program.
Course Unit Test Questions on Model Construction. The item scores on the unit
summative evaluation, for the specific questions about model construction, were an additional
source of data to make more robust the evidence produced by the rubric scores during the
modeling learning cycle. Each question was fashioned after the modeling prompts in the model
exercises and focused on the particular misconceptions addressed by the model constructions.
See APPENDIX F.
Reliability and Ethical Considerations
The Hawthorne effect, brought on by awareness of being in a study, or the John Henry
effect, marked by awareness of being in the control group, are potential threats to validity since
the high school already has a competitive atmosphere (McCambridge et al., 2014). Limiting the
information provided to students regarding the nature of the study and blinding the label of the
groups should have avoided the experimental errors of those specific possible effects. Ethical
issues, such as acquiring permission of all stakeholders and ensuring the appropriateness of the
teaching strategy was addressed through the ethics review process by Institution Review Boards
(IRBs). Approvals were obtained from the IRBs from the county school district, the high school
administration, and the higher education institution prior to the implementation of the study.
Informed consent forms and students assent forms were collected prior to the study (see
Appendix XX??). All participant data was de-identified through a number assignment to students
to use throughout the unit as part of ethical considerations.
The anticipated value of the teaching strategy was not limited to the experimental group.
The metacognitive companion to modeling was introduced to the control group during activities
later in the semester after the data collection for the research was complete. Thus, the potential
benefit of the metacognitive instruction was made possible for all students participating in the
study.
Procedures
For this study, the vehicle for eliciting experiences in metacognition within the chemistry
classroom was for students to construct science particulate models in response to their
observation of chemical phenomenon. In the quasi-experimental treatment, one class of students,
the treatment group, were provided the direct support of metacognitive instruction prior to and
during the modeling exercise, and two other classes comprising the control group, did not receive
the support of direct instruction and guidance in metacognition.

Throughout the unit of study in solution chemistry, students in both the treatment group
and the control group constructed models by drawing a particulate representation of a process or
structure in response to a phenomenon demonstrating solution chemistry. During the
construction of each model, students in the treatment group were guided towards reflecting on
metacognition, as applied to the modeling activity in chemistry, when they completed the
Metacognitive Map.
Procedures of Implementation
Length of Study. Lavi et al. (2019) recommend that a metacognitive assignment include
aspects of knowledge (MK) and regulation of cognition (MS), group work and individual work,
and a learning cycle that builds on the first metacognitive experience and transfers the
metacognitive skill into a real-life situation. The design of the current study reflected an effort to
accomplish these aspects. The study spanned over three weeks (15 school days) with research
activities scheduled for ten of the 15 instructional days. The ten days were spread out over a
three-week unit on solutions.
Pretest Administration. Before the treatment, students were given a pre-test on the
chemistry misconceptions relating to the demonstrations of the chemical phenomenon. On Day
1, chemistry content pre-test was administered for both groups.
Treatment of Metacognitive Instruction. The metacognitive instruction received by the
experimental group was via a PowerPoint introduction to metacognition and by use of a
reflective question guide, also known as the Metacognitive Map, which was completed in
tandem with the modeling activity. Students in the treatment group were presented with the
concept of metacognition prior to the unit. The introduction to metacognition presentation was
patterned after a PowerPoint presentation featured in the research of Cook et al. (2013); this
study served as the precursor for this design of research because it documented the impact of a
50-minute lecture about metacognitive learning strategies on subsequent test performance for the
experimental group (Cook et al., 2013). The PowerPoint presentation may be viewed in
APPENDIX I. The question guide, entitled Metacognition Map, was designed to stimulate
metacognitive processes of planning, regulating and evaluating and it recorded direct feedback
from the students regarding the process of modeling during the modeling exercises.
Model Construction within the Unit. Students observed chemical phenomena as the
springboard for their model construction. Students first constructed one model within a
cooperative group and transitioned to independent work, five individual efforts. The cooperative
groups used in the beginning of the study were guided by the format of schooling: all virtual
students comprised a group and F2F students were divided in one to two groups.
The phenomena were predominantly physical change demonstrations illustrating concepts in
solution chemistry. The modeling activities were embedded in an interactive digital format built
in a platform, subscribed to by the school, called Formative. There were five Formative activity
sets prepared with different types of questions, short videos and interactive sections. The group
model and the five individual models were embedded into the Formatives. Each model was
designed to address specific misconceptions and thus had specific pretest/posttest questions tied
to the content explored by the model. See APPENDIX F.
Model representations of the different types of physical interactions when substances form
solutions had the dual effect of deepening content knowledge and developing particulate
representation via modeling. The phenomena, presented via curated videos, simulations and in-
class demonstrations, addressed different aspects of solution formation including types of

solvents, solvent-solute interactions and the effect of solutes on solvent properties. Also, the
inclusion of one chemical change phenomenon in the introductory review promoted
understanding of the contrast between physical and chemical changes. The three levels of
chemistry representation were each represented as part of the modeling experience. The macro
level was the phenomenon observed by students, the symbolic was equations or graphs or
labeling and the submicroscopic was represented by the model construction. See Table 2 for the
organization and details pertaining to the Formatives and the models.

Table 2
FORMATIVES: Solution Chemistry delivered through modeling and guided metacognition

Formative # Misconception focus/ Other Topics Macro: Sub-micro: Symbolic Pretest/ FORM FORM item:
What happens when a solute dissolves in a Phenomenon Particulate Models Posttest item: Metacognitive
solvent? How can we represent this event? Items Graded Map use
1 Dissolving VS Melting Demo: sugar Model sugar dissolving Equation #1,2,3 # 13 # 12-13
What is a dissolving
Types of mixtures
solution? (video)
Covalent compounds remain intact and are
surrounded by water molecules
Characteristics of solutions
2 Ionic pairing vs dissociation when dissolved In-class DEMO: Model ionic Equation #4,5,6 # 12 #11-12
How are Adding ionic salt compound dissolving
solutions Transfer of ionic understanding from NaCl to to water
formed? KBr
Ions produced by ionic compounds equal In-class DEMO: Model covalent and Describe #7,8 #16 #15-16
subscript number in formula. Electrolytes are Adding ionic ionic compounds macroscopic
both + and - particles salts to water dissolving Equation
Polarity effect on solutions #9,10
Miscible/ soluble
Ionic compounds dissociate when dissolved
3 Increase rate does not equal increase in Demo: Solute Model different Equation #13 #12-13
Why are solubility dissolving in surface area
solutions Rate of dissolution and Effects on it Solvation different dissolving for sugar
formed? steps conditions solution
Solvation energy (Video)
4 Saturated solutions will have undissolved Saturated and Model unsaturated Solubility #11,12 # 11 #10-11
How much solute settle at bottom as paired ions and unsaturated and saturated using Curve Graph
solute can be dissolved ions in solution; unsaturated have solutions made amounts and Equation
dissolved? nothing at bottom. (Video) solubility curve graph
Temp inverse to solubility of gas #13,14,

Solubility of solids and gases 15,16
Supersaturation
5 Number of particles affect Colligative Animated vapor Model vapor pressure Labeling pure #17,18, #15 #14-15
What are properties pressure for contrasting pure substance vs. 19, 20
effects of pure solvent and solvent, covalent solution
solute? Vapor pressure is produced by solvent for solution solute and ionic
molecules above the solution/liquid solvent #21, 22
Colligative properties
Boiling and vapor pressure
Cycle of Model Skill Development. The cycle of model skill development introduced one
physical change phenomenon demonstration to model within the scope of solution chemistry on
the initial day and the next day followed with whole group critique and revision. The students
could volunteer to have the class focus on their model and this was accomplished by viewing the
model within Formative as a class. Students were guided to share strengths and recommended
additions of the model under review. With the feedback fresh, students had time to revise their
own models. The revised model was submitted for scoring on the second day. The ten days of
activity included in the research to extend over a fifteen-day unit allowed additional time for
students to internalize the practice of modeling.
Students in the experimental group completed a Metacognitive Map for each of five
individual models submitted for scoring. The experimental group had one additional instruction
on each model to complete the Metacognitive Map and to upload it to TEAMS. Use of the
Metacognitive Map each time, for individual models, was to strengthen the practice of new
metacognitive skills.
Performance Data and Survey Data Collection. At the conclusion of the unit, a post-test
on the misconceptions was administered to both groups and an administration of the subscale
two of the MCAPI was given to students in the experimental group and to students in the control
group; subscale one of the MCAPI was given to students in the experimental group. All models
created were evaluated based on the rubric and scores classified by student number and by
group.
Study Plan. The study plan is detailed in Table 3 with specific events, designated
demonstrations and instruments used. For each day of the unit, the students were in hybrid
instruction, some in face-to face instruction and some included via ZOOM in virtual setting, with
normal period time slots. See Table 3.
Table 3
Study Plan
DAY Event/ Activity
1 Administer Pre-test of content to control and experimental groups
Show Metacognitive Presentation via PPT to experimental group

2 Lecture and Problem Practice
3 Formative 1
Introductory modeling exercise in cooperative groups:
Modeling the physical change of melting
Phenomena: Sugar melting in a spoon and then burning
4 Students critique and revise group models

Formative 1
Modeling exercise independent work:
Modeling dissolution of covalent substances
Phenomena: Dissolving of sugar, C12H22O11, in water
Experimental group use Metacognitive MAP
5 Class critique and revision of individual models #1; submit model for scoring
Formative 2
Modeling dissolution of ionic substances
Phenomena: Dissolving of salt, NaBr, in water
Experimental group use Metacognitive MAP
Formative 2
Modeling dissolution of both ionic and covalent substances
Phenomena: Dissolving of KCl, K3PO4, Fructose C6H12O6
Experimental group use Metacognitive MAP individually
Formative 3 (this model not included in rubric scoring nor data for study)
Modeling dissolution in different conditions and rate of solvation
Phenomena: Dissolving of salt in different conditions of temperature and

agitation
9 Concentrations Lab
10 Formative 4
Saturated and unsaturated solutions
Phenomena: Dissolving of solute in water to make saturated and unsaturated
solutions
Formative 5
Modeling colligative properties
Phenomena: Animated vapor pressure for pure solvent and solution
Test Review
14 Unit Test
15 ADMINISTER
1. Post-test of content to control and experimental groups
2. Survey Instrument to control and experimental group
Data Analysis
All data were entered into SPSS software for screening and statistical analysis. As shown
in the research alignment table (see Table 4), two statistical methods were used to analyze
different data sets including independent sample t-tests and Pearson’s correlations. To answer
research question one (Does direct coaching or instruction of metacognition enhance modeling
and the acquisition of the conceptual content?), bivariate analysis was conducted to compare
experimental and control groups and to examine differences between the experimental group
experiencing the metacognition instruction and the control group not experiencing it. The
independent variable was the exposure to the metacognition presentation and guidance, and the
dependent variables were the rubric scores for the model and test scores. Pre-test and post-test
means were evaluated using independent sample t-test to compare learning gains. The
normalized learning gain (Hake, 1889) is considered a rough measure of effectiveness of
instruction in enhancing conceptual understanding and is utilized standardly for reporting scores
on concept inventories in research. The pretest/posttest falls in this category as a measure on
misconceptions in solution chemistry. Learning gains are described as the amount students
learned divided by the amount they could have learned (McKagan et al., 2017). The normalized
gain scores is calculated as g = (posttest – pretest) / (100 – pretest) (Hake, 1998).
To answer research question two (Does direct coaching or instruction of metacognition
enhance student awareness of metacognitive thinking processes?), bivariate analysis was
conducted to compare experimental and control groups for the MCAPI Subscale- Metacognitive
Thinking to examine differences between the experimental group experiencing the metacognition
instruction and the control group not experiencing it. To answer research question three (Is there
a correlation between student perception of values of metacognitive activities and their
awareness of metacognitive thinking processes?), Pearson Correlation was used to explore the
correlation of MCAPI-Value and MCAPI- Metacognitive Thinking between the student
perception of the value of metacognitive activities and student perception of metacognitive
thinking processes for the experimental group. See Table 4 for a summary of research questions
and analysis methodologies.

Table 4 Research Alignment

Research Question Key Variables Source of Sample Items Scale Measurement Data
Instrument(s) or and Type Analysis
Measurement of the Reliability (Cronbach’s Approach
Key Variables alpha)
(1) Does direct coaching or Students’ science Rubric for the Draw a particulate model 1-24 Continuous Independent
instruction of metacognition modeling skills Evaluation of Student depicting the solvation of the Variable Sample t test
enhance modeling and the Generated Particulate ionic salt KBr in water.
acquisition of the conceptual Models for both Reliability established locally
content? groups at the school by chemistry
instructors and by scoring 5% 0-4 Continuous Independent
of models and by another Variable Sample t test
teacher
Model construction Create a model to represent

Continuous Independent
items on unit test for aqueous solutions of .10 M 1-100
Variable Sample t test
both groups AlBr3 and .20 M C12H22O11.
Student’s science Learning gains from The process of solute particles

content acquisition misconception pre- being surrounded by solvent
test and post-test particles is known as ____.
scores for both groups a. melting
b. solvation
c. fusion
d. dehydration
Reliability established
through sources of national
tests
(2) Does direct coaching or Student awareness MCAPI, Subscale 2- Metacognitive 1-6 Interval Independent
instruction of metacognition of metacognitive Metacognitive Thinking (22 items) Likert- Variable with Sample t test
enhance student awareness thinking processes Activities & ‘I plan how to solve a type the Construct
of metacognitive thinking Perception Inventory problem before I actually start scale Mean Score
processes? based on MCAI solving the problem’
(Cooper & Sandi- High reliability shown by
Urena, 2009) for both Cronbach’s coefficient
groups All items α > .898
(3) Is there a correlation Student perception MCAPI, Subscale 1- Values (6 items) 1-6 Interval Pearson
between student perception of values of Metacognitive ‘I think using metacognitive Likert- Variable with Correlation of
of values of metacognitive metacognitive Activities & strategies could help me to type the Construct MCAPI-
activities and their activities Perception Inventory learn.’ scale Mean Score Value and
awareness of metacognitive based on MCAI Reliability shown by MCAPI-
thinking (Cooper & Sandi- Cronbach’s coefficient Metacognitive
processes? Urena, 2009) for All items α > .801 Thinking
experimental group r values
α = 0.05
Anticipated Results
Effect of Direct Instruction of Metacognition on Modeling
Data did undergo bivariate analysis to compare experimental and control groups for on
modeling rubric scores for the model to examine differences between the experimental group
experiencing the metacognition instruction and the control group. The independent variable was
exposure to the metacognition presentation and guidance, and the dependent variable was the
rubric scores. SPSS was used to enter data and to determine differences between the treatment
group, receiving the presentation of metacognition, and the control group on the performance of
the models. Means of the two groups were compared through a t-test.
Hypotheses
H : Modeling scores for control group = Modeling scores for experimental group
o
H : Modeling scores for control group ≠ Modeling scores for experimental group
a
α = 0.05
Effect of Direct Instruction of Metacognition and Modeling on Learning Gains
Pre-test and post-test data of both groups was compared to examine the possible effect of
modeling and metacognition on learning gains. The independent variable was exposure to the
metacognition presentation and guidance and the dependent variable was the normalized gain
scores calculated as g = (posttest – pretest) / (100 – pretest). An independent samples t-test was
conducted to determine a ρ value to validate support of a conclusion.
Hypotheses
H : Mean of normalized learning gain for control group = Mean of normalized learning
o
gain for experimental group
H : Mean of normalized learning gain for control group < Mean of normalized learning
a
gain for experimental group
α = 0.05
Effect of Direct Instruction of Metacognition on Metacognitive Thinking Processes
The data from the post survey MCAPI - metacognitive thinking processes subscale was
compared across groups to examine the possible effect of direct instruction and coaching of
metacognition on student perceptions of metacognitive thinking processes. The independent
variable was the use of direct instruction and coaching of metacognition to deepen practice of
metacognition and the dependent variable was the MCAPI scores. An independent sample t-test
was conducted to determine a ρ value to validate support of a conclusion.
Hypotheses
H : Mean of Survey for Control Group = Mean of Survey for Experimental Group
o
H : Mean of Survey for Control Group < Mean of Survey for Experimental Group
a
α = 0.05
Correlation between MCAPI Subscales of Experimental group

Bivariate analysis was also conducted between the two continuous variables of the
subscale MCAPI-Value and subscale MCAPI- Metacognitive Thinking of the experimental
group to determine the Pearson’s r correlation coefficient through SPSS analysis.
Hypotheses
H:ρ=ο
o
H :ρ≠ο
α
α = 0.05
Chapter Four: Results
Confirming Equivalence Before Intervention
The data was collected from a sample of honors chemistry students on their scores for the
pretest assessment (APPENDIX B) based on misconceptions in solution chemistry. For the
pretest, the mean score for the control group (n=26) was 45.12 with a standard deviation of 13.50
and the mean score for the experimental group (n=21) was 41.95 with a standard deviation of
14.26.
Pretest scores were normally distributed for both groups, as assessed by Shapiro-Wilks
test (p> .05) and that there was homogeneity of variance as assessed by Levene's Test for
Equality of Variances (p=.99). Therefore, an independent t-test was run on the data with a 95%
confidence interval (CI) for the mean difference.
Table 5
Independent Sample T-Test of Mean Comparison for Pretest Scores
Group Control Experimental t (45) p Cohen’s d
(27) (22)
Mean SD Mean SD
45.12 13.50 41.95 14.26 .78 >0.05 .23
When compared the two pretest means in our sample, the independent-sample t-test was
shown to be not statistically significant (t= .78, df =45, p>.05, one-tailed). The insignificant
difference between control and experimental groups indicated the equivalence is confirmed
before intervention was implemented. The groups are similar in size and equivalence is
confirmed before comparisons between the two groups are made in subsequent analysis.
Research Question One: Metacognition Enhancement
Independent sample t-tests were used to answer the research question one: Does direct
coaching or instruction of metacognition enhance modeling and the acquisition of the conceptual
content?
Metacognition Enhancement of Modeling
The research question was answered by examining the 1) rubric scores and 2) summative
unit test between the control and experimental groups. Peer scoring of 5% of the models scored
was implemented to fortify the reliability of the rubric use. Inter-rater reliability was established
at 42% agreement rate and confirmed with Cohen’s Kappa, K = .34 (95% CI) p < .001, which
represents a fair strength of agreement (McHugh, 2012).
The modeling rubric overall mean score for the control group (n=27) was 19.55 with a
standard deviation of 1.96 and the modeling rubric overall mean score for the experimental
group (n=22) was 20.84 with a standard deviation of 1.68. Rubric means scores were normally
distributed for both groups, as assessed by Shapiro-Wilks test (p> .05) and that there was
homogeneity of variance as assessed by Levene's Test for Equality of Variances (p=.53).
Therefore, an independent t-test was run on the data with a 95% confidence interval (CI) for the
mean difference. See Table 6 for the analysis of performance data for modeling.
When compared these two means in our sample, the independent-sample t-test was
shown to be statistically significant (t= -2.43, df =47, p=0.0095, one-tailed). In conclusion, we
reject the null hypothesis that the modeling scores for control group will equal the modeling
scores for experimental group. The 95% confidence interval is between -2.35 and -0.22. The
effect size for this analysis (d = .70) was found to exceed Cohen’s (1988) convention for medium
effect (d = .50).
Table 6
Independent Sample T-Test of Mean Comparison for Rubric Scores
(27) (22)
Mean SD Mean SD
19.55 1.96 20.84 1.68 -2.43 0.22 .70
The rubrics were examined individually to compare means and analyze differences
between the control and experimental group performances in modeling. The modeling rubric
mean scores showed an increase for both groups from rubric one to rubric five as seen in Table
7. The means for the experimental group were higher than for the control group for each rubric
score.
Rubric One. The mean score for modeling rubric one for the control group (n=26) was
17.54 with a standard deviation of 2.47 and the mean score for the experimental group (n=22)
was 19.09 with a standard deviation of 2.83. Rubric means scores were normally distributed for
both groups, as assessed by Shapiro-Wilks test (p> .05) and that there was homogeneity of
variance as assessed by Levene's Test for Equality of Variances (p=.54). Therefore, an
independent t-test was run on the data with a 95% confidence interval (CI) for the mean
difference. When compared these two means in our sample for rubric one, the independent-
sample t-test was shown to be statistically significant (t= -2.03, df =46, p=0.024, one-tailed). In
conclusion, we reject the null hypothesis that the modeling scores for control group will equal
the modeling scores for experimental group. 95% confidence interval is between -3.09 and -
0.014. The effect size for this analysis (d = .59) was found to exceed Cohen’s (1988) convention
for medium effect (d = .50).
Table 7
Independent Sample T-Test of Mean Comparison for Rubric Scores
Variable n M SD t p Cohen’s d
Rubric 1 -2.03 <0.05* .59
Control 26 17.54 2.47
Experimental 22 19.09 2.83
Rubric 2 -.128 >0.05 .037
Control 27 18.74 18.86
Rubric 3 -2.03 >0.05 .84
Control 26 19.77 20.68
Rubric 4 -3.16 <0.01** .91
Control 26 19.89 22.41
Rubric 5 -1.81 <0.05* .52
Control 26 21.85 2.28
* significance of correlation ρ < .05
** significance of correlation ρ < .01
Rubric Two. The mean score of modeling rubric two for the control group (n=27) was 18.74
with a standard deviation of 3.00 and the mean score for the experimental group (n=22) was
18.86 with a standard deviation of 3.72. Rubric means scores were normally distributed for both
groups, as assessed by Shapiro-Wilks test (p> .05) and that there was homogeneity of variance as
assessed by Levene's Test for Equality of Variances (p=.13). Therefore, an independent t-test
was run on the data with a 95% confidence interval (CI) for the mean difference. When
compared these two means in our sample for rubric two, the independent-sample t-test was
shown to be not statistically significant (t= -.128, df =47, p=0.45, one-tailed). In conclusion, we
do not reject the null hypothesis that the modeling scores for control group will equal the
modeling scores for experimental group. 95% confidence interval is between -2.05 and 1.81. The
effect size for this analysis (d = .037) was found below Cohen’s (1988) convention for small
effect (d = .037).
Rubric Three. The mean score of modeling rubric three for the control group (n=26) was
was 20.68 with a standard deviation of 2.06. Rubric means scores were normally distributed for
both groups, as assessed by Shapiro-Wilks test (p> .05) and that there was homogeneity of
variance as assessed by Levene's Test for Equality of Variances (p=.06). Therefore, an
independent t-test was run on the data with a 95% confidence interval (CI) for the mean
difference. When compared these two means in our sample for rubric three, the independent-
sample t-test was not shown to be statistically significant (t= -1.17, df =46, p=0.13, one-tailed).
In conclusion, this result does not support rejection of the null hypothesis that the modeling
scores for control group will equal the modeling scores for experimental group. 95% confidence
interval is between -2.49 and .662. The effect size for this analysis (d = .84) was found to exceed
Cohen’s (1988) convention for large effect (d = .80).
Rubric Four. The mean score of modeling rubric four for the control group (n=27) was
was 22.41 with a standard deviation of 2.50. Rubric means scores were not assessed as normally
distributed for both groups, as assessed by Shapiro-Wilks test (p< .05), and there was
homogeneity of variance as assessed by Levene's Test for Equality of Variances (p=.44).

Although the assumption of the t-test is normal distribution for dependent variables, it is
considered a fairly reliable measure for nonnormal, mound shaped distributions and is robust
even if the assumption is violated (Coolidge, 2013, p. 227). Additionally, parametric analysis for
groups means, such as t-tests, is robust even though normality assumption is violated when each
group has more than fifteen members and the group has considerably equal sample size within 1
to 1.5 ratio (Rasch et al., 2007). Therefore, an independent t-test was conducted on the data with
a 95% confidence interval (CI) for the mean difference. When compared these two means in our
sample for rubric four, the independent-sample t-test was shown to be statistically significant (t=
-3.16, df =47, p=0.001, one-tailed). In conclusion, we reject the null hypothesis that the modeling
scores for control group will equal the modeling scores for experimental group. 95% confidence
interval is between -2.49 and .662. The effect size for this analysis (d = .91) was found to exceed
Cohen’s (1988) convention for large effect (d = .80).
Rubric Five. The modeling rubric five mean score for the control group (n=27) was
21.85 with a standard deviation of 2.28 and the modeling rubric average mean score for the
experimental group (n=22) was 23.14 with a standard deviation of 2.70. Rubric means scores
were not assessed as normally distributed for both groups, as assessed by Shapiro-Wilks test (p<
.05) and that there was homogeneity of variance as assessed by Levene's Test for Equality of
Variances (p=.62). Although the assumption of the t test is normal distribution for dependent
variables, it is considered a fairly reliable measure for nonnormal, mound shaped distributions
and is robust even if the assumption is violated (Coolidge, 2013, p. 227). Additionally,
parametric analysis for groups means, such as t-tests, is robust even though normality
assumption is violated when each group has more than fifteen members and the group has
considerably equal sample size within 1 to 1.5 ratio (Rasch et al., 2007). Therefore, an
independent t-test was conducted on the data with a 95% confidence interval (CI) for the mean
difference. When compared these two means in our sample for rubric five, the independent-
sample t-test was shown to be statistically significant (t= -1.81, df =47, p=0.039, one-tailed). In
conclusion, we reject the null hypothesis that the modeling scores for control group will equal
the modeling scores for experimental group. 95% confidence interval is between -1.29 and .711.
The effect size for this analysis (d = .52) was found to exceed Cohen’s (1988) convention for
medium effect (d = .50).
Samples of students’ model construction may be viewed from Appendix K. Samples
from all classes are provided. The models are labeled for control group and experimental group.
Each of the five assignments graded by the rubric are represented in the examples shown.
Additional Findings. The data was collected from the same sample of students for the
scores on three modeling questions evaluated on the summative unit test for solutions and the
scores were averaged. The summative model construction items mean score for the control group
(n=27) was 3.22 with a standard deviation of .64 and for the experimental group (n=22) was 3.57
with a standard deviation of 0.44. Summative item scores were normally distributed for both
groups, as assessed by Shapiro-Wilks test (p> .05) and that there was homogeneity of variance as
assessed by Levene's Test for Equality of Variances (p=.11). Therefore, an independent t-test
was run on the data with a 95% confidence interval (CI) for the mean difference. See Table 8 for
the analysis of performance data for modeling on unit summative.
When compared these two means in our sample, the independent sample t-test was shown
to be statistically significant (t=-2.21, df = 47, p= 0.016, one-tailed). In, conclusion, the mean
score difference for the control and experimental group were statistically significant. 95%
confidence interval is between -0.67 and -.031. The effect size for this analysis (d = 0.55) was
found to exceed Cohen’s (1988) convention for medium effect (d = .50).
Table 8
Independent Sample T-Test and Mean Comparison for Model Construction on Unit Test
Test Items -2.21 <0.05* .63
Control 27 3.22 .64
Experimental 22 3.57 .44
Test Item 8
Control 27 3.09
Experimental 22 3.76
Test Item 15
Control 27 3.51
Test Item 24
Control 27 3.17
Metacognition Enhancement of Content Acquisition
Posttest on misconceptions in solution chemistry was collected from both control and
experimental groups. The mean score for the control group (n=26) was 77.50 with a standard
deviation of 12.86 and the mean score for the experimental group (n=22) was 80.23 with a
standard deviation of 15.63. Posttest scores were normally distributed for both groups, as
assessed by Shapiro-Wilks test (p> .05) and that there was homogeneity of variance as assessed
by Levene's Test for Equality of Variances (p=.33). Therefore, an independent t-test was run on
the data with a 95% confidence interval (CI) for the mean difference. See Table 9 for the analysis
of performance data for posttest.

Table 9
Independent Sample T-Test of Mean Comparison for Posttest
(27) (22)
Mean SD Mean SD
77.50 12.86 80.23 15.32 -.66 >0.05 .76

When compared the two posttest means in our sample, the independent-sample t-test was
shown to be not statistically significant (t= -.66, df =46, p>.05, one-tailed). The scores for the
pretest and posttest for the control and experimental group were not significantly unequal. The
effect size for this analysis (d = 0.76) was found to exceed Cohen’s (1988) convention for
medium effect (d = .50).
The data was further analyzed on their normalized learning gains from a pretest and
posttest based on misconceptions in solution chemistry. The normalized learning gain (Hake,
1998) is calculated as g = (posttest – pretest) / (100 – pretest). The pre and posttest scores for
both groups along with learning gains score are reported in Table 9.
Learning gain scores were normally distributed for both groups, as assessed by Shapiro-
Wilks test (p> .05) and that there was homogeneity of variance as assessed by Levene's Test for
Equality of Variances (p=.31). Therefore, an independent t-test was run on the data with a 95%
confidence interval (CI) for the mean difference. See Table 10 for the analysis of normalized
learning gains data for pretest/posttest.
The normalized learning gains for both groups, gexperimental = 0.65 and gcontrol = 0.60,
indicate a medium learning gain since the value falls in the range 0.7 > g > 0.3 (Pentecost &
Barbera, 2013). However, the mean learning gain score for the control group (n=25) was 0.60
with a standard deviation of 2.25 and the learning gain mean score for the experimental group
(n=21) was .65 with a standard deviation of 0.27. When compared these two means in our
sample, the independent-sample t-test was shown to be not statistically significant (t= -.676, df
=44, p>.05, one-tailed). The effect size for this analysis (d = .20) was found to equal Cohen’s
(1988) convention for small effect (d = .20). In conclusion, we fail to reject the null hypothesis
that the learning gains for control group will equal the learning gains for experimental group.
95% confidence interval is between -0.19 and 0.097.
Table 10
Normalized Learning Gains Analysis for Solutions Chemistry Pretest/Posttest
Learning Gains -.68 >0.05 .20
Control 25 .60 .22
Experimental 21 .65 .27
Pretest scores .78 >0.05 .23
Control 25 45.12 13.50
Posttest scores -.66 >0.05 .76
Control 25 77.50 12.86
Research Question Two: Direct Instruction of and Awareness of Metacognition
Independent sample t-test were used to answer the research question two: Does direct
coaching or instruction of metacognition enhance student awareness of metacognitive thinking
processes? The data was collected from a sample of honors chemistry students on their
perception of metacognitive activities in modeling contributing to metacognitive thinking

processes through MCAPI Survey. The survey subscale on metacognitive thinking processes
mean score for the control group (n=26) was 5.02 with a standard deviation of 0.67 and the
survey subscale on metacognitive thinking processes mean score for the experimental group
(n=21) was 4.92 with a standard deviation of 0.45. Homogeneity of variance was assessed by
Levene's Test for Equality of Variances (p=.26). A Shapiro – Wilks test showed a significant
departure from normality for the control group, W (26) = .91, p= .03 although the normality
assumption is only needed for small sample sizes of n < 20. For larger sample sizes, the sampling
distribution of mean is always normal regardless of how the values are distributed according to
central limit theorem (Shapiro -Wilks Test, n.d.). A Shapiro – Wilks test showed normality for
the experimental group, W (21) = .94, p= .25. Therefore, an independent t-test was conducted on
the data with a 95% confidence interval (CI) for the mean difference. See Table 11 for the
analysis of metacognitive thinking skills by independent sample t-test.
Table 11
Metacognitive Thinking Skills on MCAPI Analysis
(26) (21)
Mean SD Mean SD
5.02 .68 4.92 .45 .57 >0.05 .59
When compared these two means in our sample, the independent-sample t-test was
shown to be not statistically significant (t= .57, df =45, p>.05, one-tailed). The effect size for this
analysis (d = .59) was found to exceed Cohen’s (1988) convention for medium effect (d = .50).
In conclusion, we fail to reject the null hypothesis that the mean of metacognitive
thinking processes for control group for control group will equal to the for experimental group.
95% confidence interval is between -0.25 and 0.44. In other words, there is no significant
difference on student awareness of their metacognitive thinking processes shown in the study
between control and experimental groups.
Research Question Three: Value and Awareness Correlation in Experimental Group
Pearson’s correlation was used to answer research question three: Is there a correlation
between student perception of values of metacognitive activities and their awareness of
The data was collected from a sample of honors chemistry students in the experimental
group on their perception of value of metacognition and their perception of metacognitive
activities in modeling contributing to metacognitive thinking processes through MCAPI. The
subscale on value of metacognitive activities mean score for the experimental group (n=22) was
4.64 with a standard deviation of 0.76 and the subscale on metacognitive thinking processes
mean score for the experimental group (n=21) was 4.92 with a standard deviation of 0.45. When
examined these two means in our sample for correlation, results of the Pearson correlation
indicated that there was a significant positive association between student perception of value of
metacognition and perception of metacognitive thinking processes, (r (21) = 0.48, ρ = .015). See
Table 12.
According to Cohen’s d (Cohen, 1992), the correlational relationship is moderate in
strength (r = .48). In conclusion, we reject the null hypothesis that correlation ρ = 0. In other
words, there exists a moderate positive relationship between the student value of metacognition
and the student awareness of metacognitive thinking skills.

Table 12
Summary of Correlation Coefficients for Experimental group MCAPI (N=21)
Variables Mean (SD) Correlation with Correlation with
Value Q1-Q6 Value Q1-Q6
Pearson Significance
Correlation
Value 4.64 (.76)
Metacognitive Thinking Processes 4.92 (.45) 0.48* ρ = 0.015

Chapter Five: Discussion
The purpose of the study was to examine the interplay between science modeling and
direct instruction in metacognition. The quasi-experimental study was designed to
investigate how students’ skill in science modeling was bolstered by the fortification of
increased metacognitive skill and how each strategy, science modeling and metacognition,
augmented the acquisition of chemistry content. Further, perceptions of metacognitive value
and of metacognitive thinking processes were appraised in conjunction with the fortification of
metacognitive skill due to direct metacognitive cultivation. The goal of the study was to address
the following research questions:
1. Does direct coaching or instruction of metacognition enhance modeling and
the acquisition of the conceptual content?
3. Is there a correlation between student perception of the value of metacognitive
activities and their awareness of metacognitive thinking processes?
Direct Instruction of Metacognition Enhances Modeling
The cultivation of metacognitive skill by means of direct instruction in the PowerPoint
Presentation and of guidance through the Metacognitive Maps was shown to have a significant
enhancement for modeling in the experimental group. Each group demonstrated an increase in
rubric scores from rubric one to rubric five and from each rubric to the next one. Student skill in
modeling was shown to grow for each group as they progressed from one modeling task to
another and experienced the modeling cycle. The independent sample t-tests for both the model
rubric mean scores and the unit test models mean scores indicated greater mean scores for the
experimental group and the greater means were shown to be significant differences for the
experimental group by comparison to the control group. The medium effect size indicated by
Cohen’s d reveals that the students in the experimental group were more likely to have better
rubric scores. The outcomes from the statistical analysis, manifesting significant differences,
leads to findings concluding the direct cultivation of metacognition, contextualized in the
modeling skill development, enhanced the modeling skill for the experimental group.
Rubric mean scores for each rubric were analyzed and the means for the experimental
group were higher in each rubric than the control group means. However, rubrics one, four and
five indicated statistically significant differences between the control and experimental group yet
rubrics two and three were not shown to have significant differences. All rubric scores, but rubric
two, demonstrated a medium or large effect size as identified by the Cohen’s d. The
experimental group thus exhibited initial notable higher performances followed by similar
performances for rubric two and rubric three and then exhibited the higher performances again in
the final two rubrics. Models for rubric three, four and five were based on phenomena more
complex in nature although the same characteristics were evaluated for all models. The reflection
undertaken on the Metacognitive Maps which accompanied all model construction for the
experimental group was a likely bolster of the skill of the experimental group as the modeling
demands grew steeper.
The significant findings based on the modeling rubric score means, both the average of
combined rubric scores and the individual means of rubric one, four and five, support the main
thrust of the hypothesis that direct cultivation of metacognition contextualized within the science
skill of modeling will bolster the science skill. The enhancement of the growth in modeling skill
by co-joining it with metacognition was the central goal of the study. The findings based on the
three models assessed on the unit test were sought in to make more robust the initial findings of
the rubric score analysis. The analysis of the items from the unit summative test echoed the
statistically significant difference in performance for the experimental group when compared to
the control group. The conclusion that metacognitive cultivation bolstered the science skill of
modeling is warranted.
Mixed Results of Modeling Effects on Content Acquisition
The independent sample t-test for normalized learning gains did not exhibit statistically
significant differences between the experimental group and the control group. Learning gains
were measured based on the pretest/posttest mean scores. The experimental group did
demonstrate a higher mean score indicating a learning gain of 65% compared to the learning gain
of the control group of 60%. Yet the difference was not determined to be statistically significant.
The normalized learning gain is an important finding as it demonstrates the effectiveness
of the modeling exercises, which were founded on phenomenon demonstrating specific areas of
misconception, as one that produced medium learning gains. Although, it may be concluded that
the treatment of direct metacognitive cultivation for the experimental group did not produce
statistically significant differences from the control group, nonetheless, both groups experienced
learning gains considered a medium learning gain since the value falls in the range 0.7 > g >
0.3.
The pretest/posttest, founded on misconceptions documented in the literature and
specifically addressed by the modeling exercises, may not reflect overall content knowledge
gained in the unit pertaining to solution chemistry. There were other concepts in solution
chemistry that have not been identified as specific misconceptions and were evaluated as a part
of each student’s summative representation for the unit. A comparison of scores on the full
content of the solution unit reflected by the unit summative test might have provided more detail
to the overall comparison of control group and experimental group in content acquisition.
Another factor bearing on the results was that the experimental and the control groups
were both comprised of highly motivated honors students who demonstrated very similar
performances in learning gains; that was perhaps connected to shared characteristics of honors
students. It is also worth noting that since the pretest/posttest were conducted within a computer
platform for both face-to-face and virtual students, the opportunity for all students to screen shot
the questions during the pretest in preparation for the posttest made the performance scores less
dependable than in a normal school year when test documents are delivered in a more secure
format.
Unpacking Metacognitive Thinking Processes from the Survey Results
The students were surveyed at the completion of unit on their awareness or perception of
their metacognitive thinking processes. Students from the experimental group had been exposed
to the concept of metacognition and had reflected upon the MS tasks during each modeling cycle
however those activities did not seem to enhance their awareness of metacognition more than
that of the control group.
Comparison of survey results on MCAPI for the subscale two, perceptions of
metacognitive thinking processes, between experimental group and control groups did not
produce a statistically significant difference in the metacognitive thinking processes mean scores.
The mean score of the control group was higher but the differences were not statistically
significant. The conclusion from the analysis was that the two groups perceived their
metacognitive thinking processes when applied to the modeling exercises with similar responses
despite the treatment of the experimental group who were introduced to metacognition through
direct instruction and five Metacognitive Maps during the modeling exercises.
Metacognitive Thinking Processes and Awareness in Survey
However, the two groups perceived metacognitive thinking activities with mean scores of
5.0/6 and 4.9/6 indicating a strong positive awareness of their metacognitive processes. It is
likely that the students already possessed a level of awareness regarding metacognition or
perhaps the modeling activity cycle elicited metacognitive thinking processes with or without the
conscious awareness of metacognition inspired by the completion of the Metacognitive Maps.
Value and Thinking Processes Correlations in Experimental Group
The correlation analysis of the MCAPI revealed significant relationships predicted by the
research hypotheses. A strong statistically significant positive correlation for the experimental
group was seen in the relationship between subscale one, perception of value of metacognition
and subscale two, perception of metacognitive thinking processes. It can be concluded that there
is a statistically significant relationship for the experimental group between perception of value
of metacognition and perception of metacognitive thinking processes. Students who were
exposed to direct guidance in metacognition exhibited value correlated to their awareness of
metacognition.
Implications of Findings and Connections with Prior Literature
Students experienced scientific inquiry, as defined earlier, given that each lesson
originated with a question about solution chemistry and sought an answer through evidence of
the phenomenon to support the construction of an explanation in the format of a particulate
model. The model (explanation) constructed by the student was communicated through sharing
the visual representations, the models, with the class and instructor. The models underwent
critique by the class and revision to become a refined, justifiable answer to the question. The
normalized learning gain means, gexperimental = 0.65 and gcontrol = 0.60, indicated a medium
performance in the elimination of misconceptions in solution chemistry. The answers
internalized by the modeling process contributed to the learning gains. Thus, the scientific
inquiry approach to learning via observation and modeling (Louca & Zacharia, 2021) was shown
an effective pedagogy.
Students produced models consistent with the definition to explain and predict through
representation something in the natural world; their models show how something they observe
through the phenomenon works the way the way it does (White et al, 2009). Since modeling is
the construction and revision of a representation to communicate the characteristics of an event
and/or relationships between components of a system with explanatory and predictive capability
(Cooper et al., 2017; Kokkonen, 2017), the students constructed initial models, experienced
social constructivism (Colburn, 2007) when working with peers to isolate strengths and
weaknesses in their models in class and submitted final revised models for scoring. The social
constructivism in place and inquiry-based reflection through revision produced active knowledge
construction in keeping with the inquiry methodology (Akuma & Callaghan, 2018; Mupira &
Ramnarain, 2017).
Students experienced the modeling cycle supported in the literature (Kokkonen, 2017).
Components of the modeling exercises not only incorporated a cycle of observation-construct-
review-revise noted to be critical to modeling skill development (Kokkonen, 2017), the

phenomenon were curated to develop key concepts in solution chemistry (Santos & Arroio,
2016) and dispel literature-based misconceptions. Moreover, the modeling exercises fulfilled the
theoretical platform of the chemistry triplet of representation (Johnstone, 1993; Talanquer,
2011)): the macroscopic representation was presented through the phenomenon, the symbolic
representation was fulfilled with the required equations, specific labels and use of graph on one
exercise and the model construction covered the submicroscopic representations.
The effect of the modeling cycle was noted in the achievement of the student modeling
seen for both groups since both groups showed continual increase from one rubric to the
subsequent. Examples of the revised models produced from observation of the phenomenon to
explain phenomenon event at the particulate level are found in APPENDIX J.
The deliberate and careful design of the modeling exercises to correspond to
misconceptions and to include movement between the three representational levels fortified the
capacity of the exercises to increase learning of chemistry (Louca & Zacharia, 2012) in tandem
with the strengthening of metacognition (Thomas & Anderson, 2014). Metacognition, defined as
“individuals’ knowledge, control and awareness of cognition” (Thomas & McRobbie, 2013, p.
302), manifested in the study. Both groups perceived metacognitive thinking activities with mean
scores of 5.0/6 and 4.9/6 indicating a strong positive awareness of their metacognitive processes.
As students learned how to produce quality observation and precise representation in modeling,
they acquired new thinking strategies, monitored task performance and made changes to counter
feedback; each of these is an aspect of metacognition (Hacker, 2009; Zohar & David, 2008). The
direct cultivation of metacognitive skills (Cook et al., 2013) did not produce statistically
significant differences in metacognitive awareness for the experimental group.

However, the impact of the modeling exercises, experienced by both groups, on the
perception of metacognitive thinking processes may have overshadowed the influence of direct
cultivation of metacognition, the treatment only the experimental group experienced. Presenting
a challenge to chemistry students such as modeling, based on a phenomenon, and revising the
model, after critique, was an exceptional opportunity to elicit ME, precisely because it is a novel
encounter delivering cognitive imbalance (Sandi-Urena et al.,2011; Zohar & David, 2009). The
modeling activity cycle delivered an activity to draw out ME, with the potential to elicit MS
(Bowen, et al., 2017). As students interpreted a phenomenon, created a model and refined a
model, they used the judgments and estimates that comprise ME. ME has the potential to sharpen
MS as students develop and apply appropriate strategies to the novel encounter (Graham et al.,
2019; Thomas & McRobbie, 2013). The modeling cycle contextualized ME (Rahmen et al,
2010) and made powerful the potential for metacognitive growth (Crick, 2013; Zohar & David,
2009).
The positive scores for subscale two of MCAPI, perceptions of metacognitive thinking
processes, for both groups displayed similarities with no statistically significant difference. This
finding suggests the power of the modeling exercises to elicit metacognitive growth may have
outweighed the influence of the direct metacognitive cultivation treatment of the experimental
group. The modeling cycle may have benefitted the development of metacognition which could
have been revealed in the metacognitive awareness of both groups. The connection between
modeling, specifically sub-microscopic representation as constructed by the students, and
metacognition has been noted in the literature (Louca & Zacharia, 2012; Thomas & Anderson,
2012). Oher interventions in chemistry instruction have been shown to produce ME and have
demonstrated that science inquiry methods have a beneficial enhancement of metacognition
(Casselman & Atwood, 2017; Cooper & Sandi-Urena, 2009; Thomas & McRobbie, 2013)
Based on the literature, the researcher hypothesized that direct cultivation of
metacognition could result in higher performance, even if delivered over a shorter period of time,
as seen in this study. Achievement scores on tests (Casselman & Atwood, 2017; Cook et al.,
2013; Zepeda, et al., 2015) were increased in these studies utilizing a short treatment of
metacognitive cultivation. Other studies of longer length of treatment also saw gains in
performance data (Graham et al., 2019; Seraphin et al., 2012; Thomas & McRobbie, 2013).
The aim of the metacognitive cultivation, activated through the PowerPoint presentation
and the reflection prompted by the Metacognitive Maps, was the specific performance data of the
model scores. It was a novel contextualization to integrate the nurturing of metacognition with an
SEP, the science skill of modeling. The performance increase of the experimental group,
indicated by the achievement data of the model scores during the unit and the model scores on
the unit summative, was statistically significant in its difference from the control group. The
significant differences shown for the experimental group in overall model performance and in
rubrics one, four and five support the effectiveness of the treatment to bolster the skill thus
producing a robust claim of the capacity for enhancing performance engendered by direct
cultivation of metacognition contextualized into a content skill (Chick, 2013).
The potential of modeling to reveal and address content misconceptions was put forth in
literature (Cooper et al.,2017; Santos & Arroio, 2016; Sujak & Daniel, 2017), as well as, the
power of modeling to enhance understanding of chemistry (Edwards & Head, 2016; Kimberlin &
Yezierski, 2016; Okumus et al., 2019). The learning gains exhibited by both groups confirmed
the expectation of the literature that modeling would enhance learning and reduce
misconceptions.
Conceptual Framework Revisited
The conceptual framework anticipated the increase in performance for the modeling
achievement scores through the research question one. Affirmation of the portions of the
hypothesis regarding the strengthening of modeling as reflected by increased student mastery in
model construction was accomplished through the study. Research question three about the
correlation of perception of value and perception of thinking processes for metacognition was
answered with a positive relationship for the experimental group.
The conceptual framework did not address the possible dominance of the metacognitive
development produced by the modeling exercises that perhaps eclipsed the direct cultivation
treatment in enhancing learning gains and promoting metacognitive awareness. Given that both
groups underwent the modeling exercises with the result of demonstrating similar means in
learning gains scores for content misconceptions and similar means in perception of
metacognitive thinking processes, the catalyst for growth could have been the modeling
exercises. Scientific inquiry strategies, such as modeling, have been recognized in the literature
to produce metacognitive growth (Sandi-Uren et al., 2011; Thomas & McRobbie, 2013; Zohar &
David, 2009). Learning gains and perceptions of metacognition are complex and perhaps other
factors played a role in the results, but it is possible that a revision of the conceptual framework
to include the potential of the modeling exercises to affect change was validated by the study.
Limitations of the Study

The sampling procedure limited the methodology of this study. Using a convenience
sampling of the researcher’s classes rather than random sampling limited the generalizability of
the findings to the broader population of high school chemistry students. The study was limited
to a sample from one large public high school with majority high SES in the Southeastern United
States. The results of this study were limited to the small variety of participants since all were
enrolled in a specific level of chemistry, honors chemistry, and to a small sample size (N control =
26, N experimental = 22). A larger sample size could have aided in establishing statistically
significant findings in learning gains and survey subscale two.
The participants had common enrollment in honors chemistry at a highly competitive
high school so could be characterized as high achieving students possibly with well-honed skills
of metacognition already in place. The similarity of the participants limits the generalization of
results. The study was conducted during a time when a pandemic affected school procedures
since students were present in class virtually and face-to-face. This external factor affected
collection of data and social interaction.
Data was not collected on the specific demographics of the participants therefore the
description of the participants was defined as roughly equivalent to the overall population of the
school. This omission limited the specificity of the population description.
Additionally, a qualitative or mixed methods study, incorporating the data contained in
the student responses to the Metacognitive Map over the five model activities, may provide
further understanding of the experimental group in the present study and reduce the limitations of
a quantitative research methodology.

The choice of MCAI for adaptation was hindered by the lack of validation of internal
factors by factor analysis in the original survey. Also, the MCAPI adaptation was limited in
validity of subscale one questions and subscale two questions that could have been answered in
part by interviews of students regarding the instrument.
The MCAPI was not given prior to the treatment to both groups to avoid exposure to
metacognitive concepts before the treatment of direct guidance was delivered to the experimental
group. However, not collecting baseline data on metacognitive awareness of both groups limited
the comparison between groups and eliminated comparisons for both groups between pre and
post experiencing of the modeling exercises.
Not all conditions documented in previous research to affect metacognitive growth,
content acquisition or development of the SEP modeling skill were included in the present study.
Longer length of time for metacognitive cultivation is supported in the literature (Seraphin et al.,
2012; Thomas & McRobbie, 2013; Graham et al., 2019) for example. Measurement of content
acquisition by learning gains specifically tied to misconceptions may have limited the
comparison of full content acquisition in the area of solution chemistry; inclusion of a summative
evaluation comparison could have revealed an advantage conferred upon the experimental group
in achievement.
Future Research
Future research could be designed for longer treatment time for metacognitive cultivation
and applied to multiple science process skills or multiple SEPs. Secondly, the study could be
replicated with more participants and in a time of school protocols A larger, more diverse
population could contribute understanding of this topic and benefit the field of chemistry
instruction. A study could be developed to also address the impact of modeling on metacognition
in comparison to the direct cultivation of metacognition so as to answer some speculations which
arose from the findings of this study.
Implications to the High School Chemistry Classroom
The findings of the study and the experience of conducting it were of immense benefit to
the inform the instruction of the researcher. The modeling cycle implemented within the
Formative platform was so successful and should be imparted in that manner going forth. The
direct instruction of metacognition by PowerPoint and the cultivation induced by the
Metacognitive Maps could be applied to other performance measures early in the year to develop
metacognitive thinking processes. The MCAPI survey could be used to heighten awareness of
metacognitive thinking processes and value of metacognition and thus contribute to the academic
toolbox of the high school students towards achievement in arduous courses in high school and
beyond.
The value of each, modeling in chemistry and direct metacognitive instruction, was
demonstrated in this study. Contextualizing metacognitive instruction within a content skill was
shown to be powerful for the development of the skill. Additionally, the strength of modeling to
elicit ME and to lead toward development of metacognitive practices seem to be revealed in the
data although that was not a quest of the study. Modeling exhibited ability to reduce
misconceptions and increase content acquisition.
The study results recommend that metacognitive instruction be integrated into specific
content or into content skills for mutual benefit of skill and metacognitive practices. Modeling as
a practice is strongly recommended to enhance the learning of chemistry, elimination of
misconceptions and to contribute to metacognitive development.

References
Adadan, E., & Savasci, F. (2012). An analysis of 16-17-year-old students’ understanding of
solution chemistry concepts using a two-tier diagnostic instrument. International Journal
of Science Education, 34(4), 513–544.
Akaygun, S. & Jones, L. (2014). Words or pictures: A comparison of written and pictorial and
chemical equilibria. International Journal of Science Education, 36(5), 783-807.
Akuma, F. & Callaghan, R. (2019). Teaching practices linked to the implementation of inquiry-
based practical work in certain science classrooms. Journal of Research in Science
Teaching, 56, 64 – 90.
Alexander, P. & Winnie, P. (Eds.). (2006). Handbook of educational psychology (2nd ed.).
Mahwah, NJ, London, England: Lawrence Erlbaum Associates, Publishers.
American Chemical Society (n.d.) Science Olympiad former tests. Retrieved from
https://www.acs.org/content/acs/en/education/students/highschool/olympiad/pastexams.ht
ml
Barrow, L. (2006). A brief history of inquiry: from Dewey to standards. Journal of Science
Teacher Education, 17, 265-278.
Bernard, S. (2010). Science show making lessons relevant really matters. Retrieved from
https://www.edutopia.org/neuroscience-brain-based-learning-neuroplasticity.
Bowen, R., Picard, D., Verbene-Sutton, S., & Brame, J. (2017). Incorporating student design in
an HPLC lab activity promotes students metacognition and argumentation. Journal of
Chemical Education, 95, 108-115.

Brown, T. (2003). Making truth: metaphor in science. Urbana and Chicago, Illinois: University
of Illinois Press.
Çalýk, M., Ayas, A., & Ebenezer, J. (2005). A review of solution chemistry studies: insights into
students’ conceptions. Journal of Science Education & Technology, 14(1), 29–50.
Casselman, B. & Atwood, C. (2017). Improving general chemistry course performance through
online homework-based Metacognitive training. Journal of Chemical Education, 94
(12), 1811-1821.
Carpenter, S., Iveland, A., Moon, S., Hansen, A., Harlow, D., & Bianchini, J. (2019). Models are
a “metaphor in your brain”: how potential and preservice teachers understand the science
and engineering practice of modeling (2018). School Science and Mathematics, 119,
275–286.
Chandrasegaran, A., Treagust, D., & Mocerino, M. (2011). Facilitating high school students’ use
of multiple representations to describe and explain simple chemical reactions. Teaching
Science, 57(4), 13-20.
Chen, Y-C., Pan, Y-T., Hong, Z., Weng, X. & Lin, H-S. (2020). Exploring the pedagogical
features of integrating essential competencies od scientific inquiry in classroom teaching.
Research in Science and Technology Education, 38(2), 185-207.
Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences (2nd ed.). Hillsdale, NJ:
Lawrence Erlbaum Associates, Publishers
Cohen, J. (1992). A poller primer. Psychological Bulletin, 112(1), 155.

Chick, N. (2013). Metacognition. Vanderbilt University Center for Teaching. Retrieved from
Vanderbilt University Center for Teaching from https://cft.vanderbilt.edu/guides-sub-
pages/metacognition/#practice
Colburn, A. (2000). Constructivism: science education’s “grand unifying theory.” Clearing
House, 74(1), 9. https://doi-org.proxy.kennesaw.edu/10.1080/00098655.2000.11478630
College Board. (n.d.) AP Chemistry past exams. Retrieved from
https://apcentral.collegeboard.org/courses/ap-chemistry/exam/ap-chemistry-exam
Cook, E., Kennedy, E. & McGuire, S. (2013). Effect of teaching metacognitive learning
strategies on performance in general chemistry courses. Journal of Chemical Education, 90, 961-
967.
Coolidge, F. (2013). Statistics: A gentle introduction. Thousand Oaks, California: SAGE
Publications.
Cooper, M. & Sandi-Urena, S. (2009). Design and validation of an instrument to assess
metacognitive skillfulness in chemistry problem- solving. Journal of Chemical
Education, 86 (2), 240-245.
Cooper, M., Stieff, M. & DeSutter, D. (2017). Sketching the invisible to predict the visible; from
drawing to modeling in chemistry. Topics in Cognitive Science, 9, 902-920.
Cullen, D. (2015). Modeling instruction: a learning progression that makes high school
chemistry more coherent to students. Journal of Chemical Education, 92, 1269-1272.
DeLuca, V. W., & Lari, N. (2013). Developing students’ metacognitive skills in a data-rich
environment. Journal of STEM Education: Innovations & Research, 14(1), 45-55.

Deng, F. Chen, F., Chai, C. & Qian, Y. (2011). Constructivist-oriented data-logging activities in
Chinese chemistry classroom: enhancing students’ conceptual understanding and their
metacognition. Asia-Pacific Education Researcher (De La Salle University
Manila), 20(2), 207–221.
Devetak, I., Vogrinc, J., & Glažar, S. (2009). Assessing 16-year-old students’ understanding of
aqueous solution at submicroscopic level. Research in Science Education, 39(2), 157–
179.
Dewey, J. (1910). Science as subject-matter and as method. Science, 31, 121-127.
Donohue, K., Buck, G. & Akerson, V. (2020) Where’s the science? Exploring a new science
teacher educator’s theoretical and practical understandings of scientific inquiry.
International Journal of Research in Education and Science, 6(1), 1-12.
Dooley, C. (2017). Georgia debuts new standards for science and social studies. GeorgiaGOV.
Retrieved from https://georgia.gov/blog/2017-02-21/georgia-debuts-new-standards-
science-and-social-studies
Edwards, A. & Head. M. (2016). Introducing a culture of modeling to enhance conceptual
understanding in high school chemistry courses. Journal of Chemical Education, 93,
1377-1382.
Edwards, A., Cain, M. & Baxley, A. (2017). Rubric for the Evaluation of Student Generated
Particulate Models.
Engelmann, P. & Bannert, M. (2019) Fostering students' emotion regulation during learning:
Design and effects of a computer-based video training. International Journal of
Emotional Education, 11(2), p3-16.
Erduran, S., & Mugaloglu, E. (2014). Philosophy of chemistry in chemistry education: recent
trends and future directions. International Handbook of Research in History, Philosophy
and Science Teaching, 287-317.
Fergus, T., Limbers, C., & Bocksel, C. (2020). Associations between metacognitive beliefs and
test anxiety among middle school students. Translational Issues in Psychological
Science, 6(1), 70–80.
Flavell, J. (1979). Metacognition and cognitive monitoring: A new area of cognitive–
developmental inquiry. American Psychologist, 34, 906–911.
Fox, J. & Riconscente, M. (2008). Metacognition and self-regulation in James, Piaget and
Vygotsky. Education Psychology Review, 20, 373-389.
French, D. & Burrows, A. (2018). Evidence of science and engineering practices in preservice
secondary science teachers’ instructional planning. Journal of Science Education and
Technology, 27, 536-549.
Garcia-Martinez, J., & Serrano-Torregrosa, E. (2015). Chemistry education. Weinheim,
Germany: Wiley – VCH.
Gestsdottir, S. & Lerner, R. (2008). Positive development in adolescence: the development and
role of intentional self-regulation. Human Development, 51, 202–224.

Glaser, R. (1984). Education and thinking: The role of knowledge. American Psychology, 39, 93-
104.
Graham, K., Bohn-Gettler, C., & Raigoza, A. (2019). Metacognitive training in chemistry tutor
sessions increases first-year students’ self-efficacy. Journal of Chemical Education, 96,
1539-1547.
Gray, R. & Rogan -Klyve, A. (2018). Talking modelling: examining secondary science teachers’
modelling-related talk during a model-based inquiry unit. International Journal of
Science Education, 40(11), 1345-1366.
Hacker, D. (Eds.) (2009). Handbook of metacognition in education. New York and London:
Routledge.
Hake, R. (1998). Interactive-engagement versus traditional methods: a six-thousand-student
survey of mechanics test data for introductory physics courses. Am. J. Phys., 66(1), 64-
74. Retrieved October 28, 2021, from https://doi.org/10.1119/1.18809
Harrison, G., & Vallin, L. (2018). Evaluating the metacognitive awareness inventory using
empirical factor-structure evidence. Metacognition & Learning, 13(1), 15–38.
Heindl, M. (2019). Inquiry-based learning and the pre-requisite for its use in science at school: a
meta-analysis. Journey of Pedagogical Research, 3(2), 52-61.
Ilker Erturan, G., Arslan, Y., & Demirhan, G. (2014). A validity and reliability study of the
motivated strategies for learning questionnaire. Educational Sciences: Theory and
Practice, 14(3), 829–833.

Jackson, C. (2018) Validating and adapting the motivated strategies for learning questionnaire
(MSLQ) for STEM courses at an HBCU. Sage Journals 4(4), 1-16.
Johnstone, A. (1993). The development of chemistry teaching. Journal of Chemistry Education,
70(9), 701-705.
Justi, R. & van Driel, J. (2006). The use of interconnected model of teacher professional growth
for understanding the development of science teachers’ knowledge on models and
modelling. Teaching and Teacher Education, 22, 437-450.
K-12 Public Schools Report Card. (2021). The governor’s office of student achievement.
Retrieved from https://gaawards.gosa.ga.gov/analytics/saw.dll?dashboard
Kabapınar, F., Leach, J., and Scott, P. (2004). The design and evaluation of a teaching–learning
sequence addressing the solubility concept with Turkish secondary school students.
International Journal of Science Education 26, 635–652.
Karably, K., & Zabrucky, K. (2009). Children's metamemory: a review of the literature and
implications for the classroom. International Electronic Journal of Elementary
Education, 21(1), 32-52.
Kelly, R. M., Barrera, J. H., & Mohamed, S. C. (2010). An analysis of undergraduate general
chemistry students’ misconceptions of the submicroscopic level of precipitation
reactions. Journal of Chemical Education, 87(1), 113–118.
Khan, S. (2011). What’s missing in model‐based teaching. Journal of Science Teacher
Education, 22(6), 535–560.

Kim, B., Zyromski, B., Mariani, M., Lee, S. M., & Carey, J. C. (2017). Establishing the factor
structure of the 18-item version of the junior metacognitive awareness
inventory. Measurement & Evaluation in Counseling & Development, 50(1/2), 48–57.
Kimberlin, S. & Yezierski, E. (2016). Effectiveness of inquiry-based lessons using particulate
level models to develop high school student’s understanding of conceptual stoichiometry.
Journal of Chemical Education, 93, 1002-1009.
Kingir, S. & Aydemir, N. (2012). An investigation of the relationships among 11th grade
students’ attitudes toward chemistry, metacognition, and chemistry achievement.
GEFAD/GUJGEF, 32(3), 823-842.
Kirbulut, Z. (2014). Modeling the relationship between high school students’ chemistry self-
efficacy and metacognitive awareness. International Journal of Environmental and
Science Education, 9, 177-196.
Klar, D. (2000). Exploring science: The cognition and development of discovery processes.
Cambridge, MA: MIT Press.
Kokkonen, T. (2017). Models as relational categories. Science and Education, 26, 777-798.
Kuhn, D. (2001). Theory of mind, metacognition and reasoning. In H. Hartmann (ed.)
Metacognition in Learning and Instruction (pp. 301-326). Netherlands: Kluwer.
Lavi, R., Shwartz, G., & Dori, Y. J. (2019). Metacognition in chemistry education: a literature
review. Israel Journal of Chemistry, 59(6/7), 583–597.
Leenars, F., Joolingen, W. & Bollen, L. (2013). Using self-made drawings to support modelling
in science education. British Journal of Educational Technology, 44(1), 82-94.

Louca, L. & Zacharia, Z. (2012). Modeling-based learning in science education: Cognitive,
metacognitive, social, material and epistemological contributions. Educational Review,
64(4), 471-492.
McCambridge, J. Witton, J. & Elbourne, D. (2014). Systematic review of the Hawthorne effect:
New concepts are needed to study research participation effects. J Clin Epidemiol,
67(3):267-277.
McHugh M. (2012). Interrater reliability: The kappa statistic. Biochemia medica, 22(3), 276–
282.
McKagan,S., Sayre,S. & Madsen,S. (2017). Normalized gain: what is it and when and how
should I use it? PhysPort. Retrieved from
https://www.physport.org/recommendations/Entry.cfm?ID=93334.
Minitab. (2015). Choosing between a non-parametric test and a parametric test. Retrieved from
https://blog.minitab.com/en/adventures-in-statistics-2/choosing-between-a-
nonparametric-test-and-a-parametric-test.
Mupira, P. & Ramnarain, U. (2018). The effect of inquiry-based learning on the achievement
goal-orientation of grade 10 physical sciences learners at township school in South
Africa. Journal of Research in Science Teaching, 55, 810-825.
Naah, B. M., & Sanger, M. J. (2012). Student misconceptions in writing balanced equations for
dissolving ionic compounds in water. Chemistry Education Research and Practice, 13(3),
186-194.
Naah, B., & Sanger, M. (2013). Investigating students’ understanding of the dissolving
process. Journal of Science Education & Technology, 22(2), 103–112.
Nakhleh, M. B. (1992). Why some students don't learn chemistry: Chemical
misconceptions. Journal of chemical education, 69(3), 191.
National Academies of Sciences Engineering and Medicine. (2000). Inquiry and the National
Science Standards. Washington, D.C.: National Academies Press.
National Research Council. (1996). The National Science Education Standards Washington,
D.C.: National Academies Press.
National Research Council (2000) Center for Science, Mathematics, and Engineering Education.
Committee on Development of an Addendum to the National Science Education
Standards on Scientific Inquiry/ Inquiry and the National Science Education Standards: a
guide for teaching and learning. Washington, D.C.: National Academies Press.
National Research Council. (2012). A Framework for K-12 Science Education: Practices,
Crosscutting Concepts, and Core Ideas. Committee on a Conceptual Framework for New
K-12 Science Education Standards. Board on Science Education, Division of Behavioral
and Social Sciences and Education. Washington, DC: The National Academies Press.
Nelson, T., Stuart, R., Howard, C., & Crowley, M. (1999). Metacognition and clinical
psychology: a preliminary framework for research and practice. Clinical Psychology &
Psychotherapy, 6, 73–79.
New Jersey Center for Teaching and Learning (2014). Retrieved from
http://content.njctl.org/courses/science/chemistry/final-exam-4/final-exam-multiple-
choice-review-2/final-exam-multiple-choice-review-2014-03-26.docx
“NGSS Phenomenon.” The Wonder of Science, retrieved December 2, 2018 from
https://thewonderofscience.com/phenomenal/
Okumus, S., Koc, Y. & Doymus, K. (2019). Determining the effect of cooperative learning and
models on the conceptual understanding of the chemical reactions. Educational Policy
Analysis and Strategic Research, 14(3), 154-177.
O’Neil Jr., H. & Abedi, J. (1996). Reliability and validity of a state metacognitive inventory:
Potential for alternative assessment. The Journal of Educational Research, 89(4), 234.
Özden, M. (2009). Prospective science teachers’ conceptions of the solution chemistry. Journal
of Baltic Science Education, 8(2), 69-78.
Panaoura, A., & Philippou, G. (2003). The construct validity of an inventory for the
measurement of young pupils' metacognitive abilities in mathematics. International
Group for the Psychology of Mathematics Education, 3, 437–444.
Peters, E. E. (2009). Thinking like scientists: Using metacognitive prompts to develop nature of
science knowledge. Saarbrücken, Germany: Verlag Dr. Müller Aktiengesellschaft & Co.
KG.
Peters, E. & Kitsantas, A. (2010). The effect of nature of science metacognitive prompts on
science students’ content and nature of science knowledge, metacognition, and self-
regulatory efficacy. School Science & Mathematics, 110(8), 382-396.

Pinarbasi, T., & Canpolat, N. (n.d.). Students’ understanding of solution chemistry
concepts. Journal of Chemical Education, 80(11), 1328–1332
Pintrich, P., & de Groot, E. (1990). Motivated strategies for learning questionnaire. Retrieved
from PsycTESTS. doi:10.1037/t09161-000.
Queensland Science Teachers. (2000). Retrieved from
https://www.qldscienceteachers.com/junior-science/chemistry/quizzes/mixtures
Rahman, F., Jumani, N. Chaudry, M., Christi, S, & Abbasi, F. (2010). Impact of metacognitive
awareness on performance of students in chemistry. Contemporary Issues in Education
Research,3(10), 39-44.
Rasch, D., Teuscher, F., & Guiard, V. (2007). How robust are tests for two independent
samples? Journal of Statistical Planning and Inference, 137(8), 2706–2720.
Regents. (n.d.) Former tests. Retrieved from https://www.nysedregents.org/regents_sci.html
Rundgren, C-J. (2018) implementation of inquiry-based science education in different countries:
some reflections. Cultural Studies of Science Education, 13, 607-615.
Sandi‐Urena, S. Cooper, M. & Stevens, R. (2011) Enhancement of metacognition use and
awareness by means of a collaborative intervention. International Journal of Science
Education, 33(3), 323-340.
Sandi‐Urena, S. Cooper, M. & Stevens, R. (2012) Effect of cooperative problem-based lab
instruction on metacognition and problem-solving skills. Journal of Chemical Education,
89(6) p700-706.
Santos, V. & Arroio A. (2016).The representational levels: influences and contributions to
research in chemical education. Journal of Turkish Science Education, 13(13), 3-18.
Schraw, G. (1998). Promoting general metacognitive awareness. Instructional Science, 26, 113-
125.
Schraw, G., Crippen, K., & Hartley, K. (2006). Promoting self-regulation in science education:
metacognition as part of a broader perspective on learning. Research in Science
Education, 36, 111-139.
Schraw, G. & Dennison , R. (1994). Assessing metacognitive awareness. Contemporary
Educational Psychology, 19(4), 460-47
SPSS Shapiro-Wilk Test. SPSS tutorials SPSS Shapiro-Wilk Test Quick Tutorial with Example
Comments. (n.d.). Retrieved November 6, 2021, from https://www.spss-tutorials.com/spss-
shapiro-wilk-test-for-normality/.
Smith, K. J., & Metz, P. A. (1996). Evaluating student understanding of solution chemistry
through microscopic representations. Journal of Chemical Education, 73(3), 233.
Schwartz, R., Lederman, N., & Crawford, B. (2004). Developing views of nature of science in an
authentic context: An explicit approach to bridging the gap between nature of science and
scientific inquiry. Science Education, 88(4), 610-645.
Schwarz, C., Passmore, C. & Reiser, B. (2017) Helping Students Make Sense of the World Using
Next Generation Science and Engineering Practices. Arlington, VA: NSTA Press.
Schwarz, C., & White, B. (2005). Metamodeling knowledge: Developing students’
understanding of scientific modeling. Cognition and Instruction, 23(2), 165–205.

Science and Engineering Practices. (2013) NSTA. Retrieved July 13, 2020 from
http://static.nsta.org/ngss/resources/MatrixForK-
12ProgressionOfScienceAndEngineeringPracticesInNGSS.8.14.14.pdf
Smith, J. diSessa, A., & Roschlle, J. (1993). Misconceptions reconceived: A constructionist
analysis of knowledge in transition. The Journal of the Learning Sciences, 3(2), 115-163.
Sperling, R., Howard, L. & Murphy, C. (2002). Measures of children´s knowledge and
regulation of cognition, Contemporary Educational Psychology, 27, 51-79.
Sujak, K. & Daniel, E. (2017). Understanding of macroscopic, microscopic and symbolic
representations among form four students in solving stoichiometric problems. Malaysian
Online Journal of Education Sciences, 2(3), 83-96.
Talanquer, V. (2011). Macro, submicro and symbolic: the many faces of the chemistry “triplet”.
International Journal of Science Education, 32(2), 170-195.
Talanquer, V. (2018). Chemical rationales: another triplet for chemical thinking. International
Journal of Science Education, 40(15), 1874.
Tanner, K. (2012). Promoting student metacognition. CBE-Life Sciences Education, 11, 113-120.
Thomas, G. (2002). Conceptualizations, development and validation of an instrument for
investigating the metacognitive orientation of science classroom learning environments:
The metacognitive orientation learning environment scale –Science (MOLES – S).
Learning Environments Research, 6, 175–197.
Thomas, G., Anderson, D., Nashon, S. (2008). Development of an instrument designed to
investigate elements of science students' metacognition, self-efficacy and learning

processes: The SEMLI-S. International Journal of Science Education, 30(13), 1701–
1724.
Thomas, G. & Anderson, D. (2014). Changing the metacognitive orientation of a classroom
environment to enhance students’ metacognition regarding chemistry learning. Learning
Environment Res, 17, 139-155.
Thomas, G. & McRobbie, C. (2001). Using a metaphor for learning to improve students’
metacognition in the chemistry classroom. Journal of Research in Science Education,
38(92), 222-259.
Thomas, G., & McRobbie, C. (2013). Eliciting metacognitive experiences and reflection in a
year 11 chemistry classroom: an activity theory perspective. Journal of Scientific
Educational Technology, 22, 300-313.
Towns, M. (2013). New guidelines for chemistry education research manuscripts and future
directions of the field. Journal of Chemical Education, 90, 1107-1108.
Uopasai, S., Bunterm, T., Supaporn, M., & Tang, K. (2018). The effect of constructivism,
metacognition and neurocognitive-based teaching model to enhance veterinary medicine
students’ learning outcomes. Pertanik Journal of Social Sciences and Humanities, 26 (4),
2313-2331.
U.S. News and World Report. (2020). Education: Walton High School. Retrieved from
https://www.usnews.com/education/best-high-schools/georgia/districts/cobb-
county/walton-high-school-5793
Veenman, M., Van Hout-Wolters, B., & Afflerback, P. (2006). Metacognition and learning:
conceptual and methodological considerations. Metacognition Learning, 1, 3-14.
Veenman, M. (2011). Learning to Self-Monitor and Self-Regulate. In R. Mayer, & P. Alexander
(Eds.), Handbook of Research on Learning and Instruction (pp. 197-218). New York:
Routledge.
Vigeant, F. What is NGSS? A brief history. Know Atom’s blog. Retrieved from
https://www.knowatom.com/blog/what-is-ngss-a-brief-history
Ward, R. & Butler, D. (2019). An investigation of metacognitive awareness and academic
performance in college freshmen. Education, 139(3), 120-126.
White, B. & Fredericksen, J. (2005). A theoretical framework and approach for fostering
metacognitive development. Educational Psychologist, 40(4), 211-223.
Windschitl, M., Thompson, J., & Braaten, M. (2008). Beyond the scientific method: Model‐
based inquiry as a new paradigm of preference for school science investigations. Science
Education, 92(5), 941–967.
Yen, M.-H., Wang, C.-Y., Chang, W.-H., Chen, S., Hsu, Y.-S., & Liu, T.-C. (2018). Assessing
metacognitive components in self-regulated reading of science texts in e-based
environments. International Journal of Science & Mathematics Education, 16(5), 797–
816.
Zepeda, C., Hlurkowsky, C., Partika, A. & Nokes-Malach, T. (2019). Identifying teachers’
supports of metacognition through classroom talk and its relation to growth in conceptual
learning. Journal of Educational Psychology, 111(3), 522-541.

Zepeda, C., Richey, J. Ronevich, P, & Nokes-Malach, T. (2015). Direct instruction of
metacognition benefits adolescent science learning, transfer, and motivation: an in vivo
study. Journal of Educational Psychology, 107(4), 954-970.
Zohar, A. & Ben David, A. (2008). Explicit teaching of meta-strategic knowledge in authentic
classroom situations. Metacognition Learning, 3, 59-82.
Zohar, A. & Ben David, A. (2009). Paving a clear path in a thick forest: a conceptual analysis of
a metacognitive component. Metacognition Learning, 4, 177-195.
Zubeyde, D. (2014). Modeling the relationship between high school students’ chemistry self-
efficacy and metacognitive awareness. International Journal of Environmental and
Science Education, 9, 177-196.

Appendix
A.
PARENTAL CONSENT FORM WITH CHILD ASSENT STATEMENT

Title of Research Study: Metacognition and Modeling in Chemistry Instruction
Researcher's Contact Information: Meri Laird Cain Meri.Cain@cobbk12.org
Dear Parent:
Your child is being invited to take part in a research study conducted by Meri Cain of
Kennesaw State University. Before you decide to allow your child to participate in
this study, you should read this form and ask questions if you do not understand. Your
decision to participate or not participate in this research will have no bearing on your student's
grades or class standing. The study is not related to their relationship with the instructor or
school and cannot / will not affect the relationship with the instructor or with the school.
Description of Project
The purpose of the study is to investigate the benefit of directly teaching students
about metacognition and of guiding students to reflect about metacognitive
activities on the construction of chemistry models and on learning gains in
chemistry.
Explanation of Procedures
Students will have a pre-test and post-test (this is a normal part of the course work)
on solution chemistry.
Students will be considered in control groups or experimental groups based on class

sections. Students in the experimental groups will have a short presentation about
metacognition delivered using PowerPoint and will use metacognitive guides as they
work on chemistry modeling. (Students, in the control groups, who do not see the
Metacognitive Presentation and use the metacognitive guides as part of the study will
do so in the next unit ensuring all students learn about metacognition.)
Students will practice chemistry modeling based on chemical phenomenon and models
will be graded (this is a normal part of the course work). Students will take a
metacognitive survey. Data from the guides, the models, the tests and the survey will
be used in the study although names will be concealed.
Time Required
Additional tasks to regular course work will take 10 minutes on 5 different days during
a 3-week unit. Both control and experimental groups will use metacognitive guides
either in this unit, associated with the study, or in the next unit.
Risks or Discomforts
There are no known risks or discomforts.
Benefits
Students may benefit from learning about metacognition and reflecting upon their
own metacognitive experiences.
Confidentiality
The results of this participation will be confidential. Names will be concealed from all
materials used in the study. Students will be assigned a study I.D. to label all materials used in
place of a name. The study I.D. will allow for linking of data for analysis but protect the identity
of the students. All of the data will be stored in secure password protected files.
Use of Online Surveys (if applicable)

Email addresses nor I.P. Addresses will be collected.
Inclusion Criteria for Participation

Students are enrolled in the researcher’s classes and are in the grade level of 10 th –
12th grade.
Parental Consent to Participate
I give my consent for my child, ______________________________________, to

participate in the research project described above. I understand that this
participation is voluntary and that I may withdraw my consent at any time without
penalty. I also understand that my child may withdraw his/her assent at any time
without penalty.
__________________________________________________ ___/___/2021
Signature of Parent or Authorized Representative, Date
__________________________________________________ ___/___/2021
Signature of Investigator, Date
__________________________________________________________________________
PLEASE SIGN TWO COPIES OF THIS FORM, KEEP ONE AND RETURN THE OTHER TO
THE INVESTIGATOR
Research at Kennesaw State University that involves human participants is carried out
under the oversight of an Institutional Review Board. Address questions or problems
regarding these activities to the Institutional Review Board, Kennesaw State
University, 585 Cobb Avenue, KH3417, Kennesaw, GA 30144-5591, (470) 578-7721.
__________________________________________________________________________
CHILD/ Student ASSENT STATEMENT
Child Assent to Participate
As your teacher, Ms. Meri Cain, I am inviting you to be in a research study about
metacognition and modeling in chemistry. Your parent has given permission for you
to be in this study, but you get to make the final choice. It is up to you whether you
participate.
Your decision to participate or not participate in this research will have no bearing on your
grades or class standing. The study is not related to your relationship with the instructor or school
and cannot / will not affect the relationship with the instructor or with the school.
During the study, I will ask you to watch a metacognitive presentation, use your data
from completed metacognitive guides, pretest and posttest scores, model rubric
scores and your data from a metacognitive survey. See the details below:
Description of Project
The purpose of the study is to investigate the benefit of directly teaching students
about metacognition and of guiding students to reflect about metacognitive
activities on the construction of chemistry models and on learning gains in
chemistry.
Explanation of Procedures
Students will have a pre-test and post-test (this is a normal part of the course work)
on solution chemistry.
Students will be considered in control groups or experimental groups based on class

sections. Students in the experimental groups will have a short presentation about
metacognition delivered using PowerPoint and will use metacognitive guides as they
work on chemistry modeling. (Students, in the control groups, who do not see the
Metacognitive Presentation and use the metacognitive guides as part of the study will
do so in the next unit ensuring all students learn about metacognition.)
Students will practice chemistry modeling based on chemical phenomenon and models
will be graded (this is a normal part of the course work). Students will take a
metacognitive survey. Data from the guides, the models, the tests and the survey will
be used in the study although names will be concealed.
Time Required
Additional tasks to regular course work will take 10 minutes on 5 different days during
a 3-week unit. Both control and experimental groups will use metacognitive guides
either in this unit, associated with the study, or in the next unit.
Risks or Discomforts
There are no known risks or discomforts.
Benefits
Students may benefit from learning about metacognition and reflecting upon their
own metacognitive experiences.
Confidentiality
The results of this participation will be confidential. Names will be concealed from all
materials used in the study. Students will be assigned a study I.D. to label all materials used in
place of a name. The study I.D. will allow for linking of data for analysis but protect the identity
of the students. All of the data will be stored in secure password protected files.
Use of Online Surveys (if applicable)

Email addresses nor I.P. Addresses will be collected.
Inclusion Criteria for Participation

Students are enrolled in the researcher’s classes and are in the grade level of 10th –
12th grade.
Students, you do not have to answer any question you do not want to answer or do
anything that you do not want to do. Everything you say and do will be private, and
your parents will not be told what you say or do while you are taking part in the
study. When I tell other people what I learned in the study, I will not tell them your
name or the name of anyone else who took part in the research study.
If anything in the study worries you or makes you uncomfortable, let me know and you
can stop. No one will be upset with you if you change your mind and decide not to
participate. You are free to ask questions at any time and you can talk to your parent
any time you want. If you want to be in the study, sign or print your name on the line
below:
_____________________________________________ ___/___/2021
Child’s/Student’s Name and Signature, Date
Check which of the following applies (completed by person administering the assent.)
 Child/Student is capable of reading and understanding the assent form and has
signed above as documentation of assent to take part in this study.
 Child is not capable of reading the assent form, but the information was
verbally explained to him/her. The child signed above as documentation of
assent to take part in this study.
____________________________________________ ___/___/2021
Signature of Person Obtaining Assent, Date
B. Instruction in Metacognition, Metacognitive MAP M. Cain (2019)
Metacognition MAP*
thinking about thinking
Planning ● Monitoring ● Evaluating
name: ______________________________________ period: _____ date: ______/______/____________
Task: Creating and Labeling/Keying a Model to Represent a Phenomenon in Chemistry
Making a plan
What do I know about models and creating a model in science?
How could I describe the instructional goal of my teacher in having the class draw models in chemistry?
What are the steps I need to take to successfully create a model based on a phenomenon in chemistry?
Actively monitoring
What strategies am I using that are working to help me create a model? What is one strategy I have discarded?
While creating the model, what is most challenging and what is most confusing to me?
What other resource could I be using to complete this task?
Practice evaluating
To what extent did I successfully accomplish the goal of the task?
If I were the teacher, what would I identify as one strength of my work; what would I identify as one flaw in my
work?
What are the sequential steps I could take to create a model of another chemistry phenomenon?
How could I teach modeling to another student?
* based on the work of Tanner, K. (2012). Promoting student metacognition. CBE—Life Sciences Education, 11, 113-120.
C. Pretest/Posttest on Solutions Misconceptions

Solutions Misconceptions Pre-Test/Posttest
1. When sugar C12H22O11 dissolves in water, it:
a. melts into a liquid and forms a homogenous mixture with water
b. breaks apart into smaller particles, invisible to the naked eye, to form homogenous
mixture with water
c. reacts to form a completely new substance that can disappear in water
d. joins with water molecules to form a new invisible substance
(Queensland Science Teachers, 2000)
2. The process of solute particles being surrounded by solvent particles is known as _______.
e. melting
f. solvation
g. fusion
h. dehydration
(New Jersey Center for Teaching and Learning, 2014)
3. The equation that describes the process represented in the model is:
a. C12H22O11 (s) → C12H22O11 (l)

b. C12H22O11 (l) → C12H22O11 (s)
c. C12H22O11 (s) → C12H22O11 (aq)
d. C12H22O11 (s) + 12O2(g)→ 12CO2(g)+ 12H2O(l)
4. Which of the images below best represents an ionic compound like KBr dissolved in water?
(College Board, n.d.)

5. A solid compound of a group 1 (alkali) metal and a group 17 (halogen) element dissolves in
water. The diagram above represents one type of solute particle present in the solution.
Which of the following identifies the solute particle and best helps explain how the solute
particle interacts with water molecules?
a. The particle is a negative ion and water molecules surround it in a hydration ring.
b. The particle is a positive ion, and water molecules surround it in a hydration ring.
c. The particle is a ionic molecule and water molecules surround it in a hydration ring.
d. The particle is a covalent molecule and water molecules surround it in a hydration
ring.
6. Which of the diagrams above best represents the interactions that are responsible for
the relatively large solubility of KCl crystals in water, and why?
a. Diagram 1, because strong ion-dipole interactions between KCl and water help to
dissociate the solute.
b. Diagram 1, because strong London dispersion forces between the K+ and Cl− ions and
water replace the weak London dispersion forces between two water molecules.
c. Diagram 2, because strong dipole-dipole forces between KCl and water help to
separate the KCl units within the crystals.
d. Diagram 2, because the hydrogen bonds between water molecules expand to
accommodate the KCl particles and pull them into solution.
7. Which of these equations best describes what happens when solid MgCl2 dissolves in
water?
a. MgCl2(s) → MgCl2 (aq)
b. MgCl2(s) → MgCl2 (l)
c. MgCl2(s) → Mg2+(aq) + 2Cl- (aq)
d. MgCl2(s) → Mg2+(aq) + Cl-2 (aq)
(Naah & Sanger, 2012)
8. Which of these diagrams best describes what happens when solid MgCl2 dissolves in
water?
(Naah & Sanger, 2012)

9. Which statement describes an electrolyte?
a. An electrolyte conducts an electric current as a solid and dissolves in water.
b. An electrolyte conducts an electric current as a solid and does not dissolve in water.
c. When an electrolyte dissolves in water, the resulting solution conducts an electric
current.
d. When an electrolyte dissolves in water, the resulting solution does not conduct an
electric current
(Regents, n.d.)
10. Why is potassium nitrate classified as an electrolyte?
a. It is a molecular compound.
b. It contains a metal.
c. It can conduct electricity as a solid.
d. It releases positive and negative ions in an aqueous solution.
(Regents, n.d.)
11. There are different concentrations of sugar solutions in beakers A, B, and C. One
solution is saturated, one unsaturated and one supersaturated. The dots represent the
sugar molecules; water is not shown. Which is correct?
a. A is saturated; B is unsaturated; C is supersaturated

b. A is unsaturated; B is supersaturated; C is saturated
c. A is supersaturated; B is unsaturated; C is saturated
d. A is supersaturated; B is saturated; C is unsaturated
(Adadon & Savasci, 2012)
12. The solubility of KClO3 at several temperatures is shown in the accompanying diagram.
A student mixes 10.0 g of KClO3 with 45.0 g of H2O and stirs it for a long time at 60 ˚C
until the solution is completely clear then allows it to cool slowly to 20 ˚C where it
remains clear. Which statement about the final clear mixture at 20 ˚C is correct?
a. It is a saturated solution.
b. It is an unsaturated solution and can be made saturated by decreasing the
temperature.
c. It is an unsaturated solution and can be made saturated by increasing the
temperature.
d. It is a supersaturated solution.
(American Chemical Society, n.d.)
13. The solubility of gases in water:
a. is independent of the temperature.
b. increases with increasing temperature.
c. decreases with increasing temperature.
d. Gases are not soluble in water.
e. none of the above
14. Under which conditions is the solubility of oxygen gas in water the greatest?
Pressure Temperature
a. high high
b. high low
c. low high
d. low low
15. The bottler of a carbonated beverage dissolves carbon dioxide in water by placing
carbon dioxide in contact with water at a pressure of 1 atm at room temperature. The
best way to increase the amount of dissolved CO2, would be to
a. increase the temperature and increase the pressure of CO2.

b. decrease the temperature and decrease the pressure of CO2.
c. increase the temperature without changing the pressure of CO2.
d. decrease the temperature and increase the pressure of CO2.
e. increase the pressure of CO2 without changing the temperature
16. The curve best represents the solubility curve

for____at 25oC
a. LiBr
b. C12H22O11
c. CO2
d. NaI
17. For which property is the value greater for a solution

of a nonvolatile solute than for the pure solvent?
a. boiling point
b. freezing point
c. triple point
d. vapor
18. Which of the following statements about colligative properties is FALSE?
a. The boiling point of a solution is increased by the addition of salt.
b. The freezing point of a solution is lowered by the addition of salt.
c. The number of particles dissolved is not a factor.
d. The identity of the solute is not a factor.
e. All of the above statements are true
19. Compared to the boiling point and the freezing point of water at 1 atmosphere, a 1.0 M
CaCl2(aq) solution at 1 atmosphere has a
a. lower boiling point and a lower freezing point
b. lower boiling point and a higher freezing point
c. higher boiling point and a lower freezing point
d. (4) higher boiling point and higher freezing point
(Regents, n.d.)
20. Which solute produces the highest boiling point in a 0.15 m aqueous solution?
a. CaCl2
b. NaBr
c. CuSO4
d. CH3OH
21. The normal boiling point and vapor pressure at 25 °C are measured for liquids in two
flasks. Flask A contains pure water and flask B contains a 1.0 M aqueous NaCl solution.
Which flask contains the liquid with the higher boiling point?
Which flask contains the liquid with the least molecules of water vapor above the surface?
Higher boiling point Least molecules of water vapor
a. Flask A Flask A
b. Flask A Flask B
c. Flask B Flask A
d. Flask B Flask B
22. Sugar is dissolved in water as modeled in the following diagram:
Which statement is true?
a. The vapor pressure for the solution is lower since less water is escaping the solution.
b. The vapor pressure for the solution is lower since less sugar is escaping the solution.
c. The vapor pressure for the solution is higher since more water is escaping the solution.
d. The vapor pressure for the solution is lower since more sugar is escaping the solution.
(Regents, n.d.)
D. Rubric to evaluate student models

Rubric for the Evaluation of Student Generated Particulate Models A. EDWARDS, M. CAIN, A. BAXLEY (2017)
Model feature(s) 4 3 2 1
Student uses the appropriate Student uses different geometric

Student uses the chemical
representation (i.e., Lewis shapes to represent each A drawing at the macroscopic
Illustration of formula(e) of the
Structure, ball and stick, space different chemical formula in the level (no detail) is provided in
Particles molecules/atoms to represent
filling model) of each of the model. A legend/key is also student answer
the substance in the model
chemical formula(e) of the provided in the illustration
molecules/atoms in the model
Spacing between particles is Spacing in the model is same

Spacing between particles is
appropriate for two of the Spacing between particles is regardless the state of matter
appropriate for the state of
Use of spacing following regarding model: the appropriate for only one of the illustrated.
matter, energy/temperature,
And/or size state of matter, energy, following regarding model: the Energy/temperature or
and attractive forces present in
temperature, and/or attractive state of matter, attractive forces present in
the sample
forces present in the sample the sample are not addressed
Motion (vibrational,
Description of motion Motion (vibrational,
translational, rotational) is
(vibrational, translational, translational, rotational) is
appropriate for the state of
rotational) is appropriate for the appropriate for the Neither motion nor energy of
Use of matter present and is
state of matter present and is state of matter present and is particles is addressed in the
motion/energy appropriate for the
appropriate for the appropriate for the model
energy/temperature present but
energy/temperature present energy/temperature present but
is provided in words instead of as
is provided in words instead of
an aspect of the model
as an aspect of the model
Orientation of particles shows a Orientation of particles shows a Both intramolecular and Major errors or
clear understanding of the clear understanding of the bonding interactions, but inconsistencies in both
Particle forces of attraction/repulsion forces of attraction/repulsion inconsistencies or minor errors intramolecular and bonding
interaction between particles. Additionally, between particles. Minor errors are present with regards to both interactions in the model or
appropriate intramolecular and are present regarding the types of interactions the interactions between
bonding interactions are bonding interactions particles is not addressed in
illustrated correctly the illustration
An accurate description of the Some attempt at an accurate Little attempt at an accurate No attempt at an accurate
Number of number/relative amount of description of the description of the description of the
particles molecules present is depicted number/relative amount of number/relative amount of number/relative amount of
depicted based on mole ratio or molecules present is depicted molecules present is depicted molecules present is depicted
stoichiometric data based on mole ratio or based on mole ratio or based on mole ratio or
stoichiometric data stoichiometric data stoichiometric data
(at least 50% correct) (at least 25% correct)
Macroscopic aspects Many macroscopic aspects Few macroscopic aspects No macroscopic aspects
meaningful/appropriate to the meaningful/appropriate to the meaningful/appropriate to the meaningful/appropriate to
Use of levels of physical or chemical process are physical or chemical process are physical or chemical process are the
representation consistently depicted and are depicted and are accurate. depicted leading to significant physical or chemical process
accurate Omission of macroscopic aspects difficulty in using/understanding are presented in the
does not affect the overall the illustration illustration
understanding of the illustration
Complete particle Basic Particle Mixed Descriptive

*adapted from:
Merritt, J., & Krajcik, J. (2013). Learning progression developed to support students in building a particle model of matter. In Concepts of matter
in science education (pp. 11-45). Springer, Dordrecht.
Merritt, J. D., Krajcik, J., & Shwartz, Y. (2008, June). Development of a learning progression for the particle model of matter. In Proceedings of
the 8th international conference on International conference for the learning sciences-Volume 2 (pp. 75-81). International Society of the
Learning Sciences.
Ngai, C., Sevian, H., & Talanquer, V. (2014). What is this substance? What makes it different? Mapping progression in students’ assumptions
about chemical identity. International Journal of Science Education, 36(14), 2438-2461.
Sevian, H., & Talanquer, V. (2014). Rethinking chemistry: A learning progression on chemical thinking. Chemistry Education Research and
Practice, 15(1), 10-23.
E. Instructions for Modeling Exercises as appeared in Formative

Model 1
Control Group instructions: Experimental Group instructions:
Initial model template Revised Model template
Model 2: only control group instructions shown

Model 4: only experimental group instructions shown

F. Unit Summative Questions # 8, 15, 24 on Model Construction

# 8 Dissolving of covalent and ionic solutes
# 15 Saturation state of solutions using solubility table
# 24 Vapor pressure changes based on number of particles; particles in ionic solute.

G. Inventory assessing metacognition perception and activities*** M Cain (2019)
MCAPI Metacognitive Activities and Perception Inventory MCain2019

Subject: Chemistry 111 Grade Level: 11/12 Time Frame: _____
Information about you: Check ONE for each

Gender □ Male □ Female Average in course □ 95-100 □ 90-95 □ 80-89 □ 70-79 □ 60-69
Respond to each statement with the appropriate number from the response scale.
1- Strongly Disagree
2- Somewhat Disagree
3- Slightly Disagree
4- Slightly Agree
Statements 6-
5-
Somewhat Agree
Strongly agree
1 2 3 4 5 6
Perception of value of metacognitive activity and strategies

Q I was helped by the presentation to look at academic tasks differently.
1
Q I will use some of the strategies to approach my academic work differently.

2
Q I understood the ideas presented about metacognition.

3
Q I think using metacognitive strategies could help me to learn.

4
Q I have already used a new strategy in my academic tasks after working on metacognition.
5
Q I believe that the metacognitive strategies helped me with the task of modeling at the sub-
6 microscopic level
Perceptions/awareness of own metacognitive thinking process

Q I read the modeling activity instructions carefully to fully understand it and determine what
7 the goal is.
Q When I do modeling activities, I try to learn more about the concepts so I can apply this
8 knowledge to the test
Q I sort the observations I made of the phenomenon and determine what is relevant.
9
Q I try to relate the creation of my model to previous models I have created or models seen.
10
Q I clearly identify the goal of the modeling activity before attempting to create my model.
11
Q I consider what information needed might not be evident in the demonstration of the
12 chemical phenomenon.
Q I try to double-check everything: my observations of the phenomenon, ratios of particles,

13 particulate representation, appropriate symbols and keys and processes shown.
Q I use graphic organizers (diagrams, flow-chart) to better understand the chemistry concepts
14 or processes.
Q I experience moments of insight or creativity when creating chemistry models

15
Q I jot down things I know that might help me create the model or with the chemistry concept
16 before creating the model.
Q I find important relations between particles, particle quantities and particle placements
17 before creating my model
Q I make sure that my finished model actually represents the chemical phenomenon and
18 achieves the goal of the activity.
Q I plan how to create the model before I actually start drawing (even if it is a brief mental
19 plan).
Q I reflect upon things I know are relevant to the chemical phenomenon and the chemistry
20 concept.
Q I analyze the steps of my plan and the appropriateness of each step.

21
Q I attempt to break down the observation of phenomenon to find a starting point for my
22 model.
Q I spend little time on creating a model for a concept that I have not been taught.
23
Q When I create a model, I omit thinking of concept before I attempt my drawing of the
24 model.
Q Once I know how to draw the model, I put no more time in understanding the concepts
25 involved.
Q I do not check to see if my model representations make sense.

26
Q If I do not know exactly how to represent the chemical phenomenon, I immediately start to
27 draw something to turn in.
Q I start drawing my model without having read all the details of the activity and consulting
28 my observations of the phenomenon.
***Adapted from MCAI, Metacognitive Activities Inventory (Cooper & Sandi-Urena, 2009)
H. Subscales and components of MCAPI

I. Metacognition Power Point

J. Model Example Responses: *control group **experimental group

Model 1:
* * **
Model 2:
* * **
Model 3:
* * **
Model 4:
* * **
Model 5:
* * **

Metacognition and Modeling in Chemistry Instruction

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Metacognition and Modeling in Chemistry Instruction

Uploaded by

Copyright:

Available Formats

Kennesaw State University

DigitalCommons@Kennesaw State University

Doctor of Education in Secondary Education Department of Secondary and Middle Grades